ICA 2019 and EAA Euroregio

Session chair(s):

J.V. Patrício, LNEC

Sound and Noise around us Marion Anne Burgess [Abstract] Sound is fundamental to our social structure from the basic need for communication through to the expression of our emotions in music. Sound at too high a level can cause damage to hearing and it is the role of society, workplace management and governments to ensure that the community is aware of the risk and that this risk is minimised by implementing noise control and noise management. Extensive research has shown that sound levels that commonly occur in our cities and rural areas and that may be below the damage risk level can create annoyance and lead to various health effects. Sometimes we may desire silence. But generally, there is a level of noise that we consider acceptable; when the noise from outside does not interfere with what we are doing at that time. So limits and control measures need to be implemented to keep community noise within the acceptable range while still allowing for the necessary business, industry, transportation and entertainment to continue. Engineering noise control at source is a solution for some problems. For others an acceptable environment may be achieved by coupling a deep understanding of the effects of noise with the application of some of many tools available using modern technology. Promotion of best practices in the wide community is one of the goals for the activities that are part of the ICA organised International Year of Sound in 2020.

Session chair(s):

M. Kob, Erich Thienhaus Institute, Detmold University of Music

Sounds of the human vocal instrument Shrikanth Narayanan [Abstract] The vocal tract is the universal human instrument played with great dexterity to produce the elegant acoustic structuring of speech, song and other sounds to communicate intent and emotions. The sounds produced by the vocal instrument also carry crucial information about individual identity and the state of health and wellbeing. A longstanding research challenge has been in improving the understanding of how vocal tract structure and function interact, and notably in illuminating the variant and invariant aspects of speech (and beyond) within and across individuals. The first part of the talk will highlight engineering advances that allow us to perform investigations on the human vocal tract in action– from capturing the dynamics of vocal production using novel real-time magnetic resonance imaging to machine learning based articulatory-audio modeling–to offer insights about how we produce sounds with the vocal instrument. The second part of the talk will highlight some scientific, technological and clinical applications using such multimodal data driven approaches in the study of the human vocal instrument.

Session chair(s):

J. Fels, Teaching and Research Area of Medical Acoustics

Astroparticle Immersive Synthesizer³ or how cosmic “ghost particles” inspire a novel concept of spatialisation of sound Christopher Wiebusch, Tim Otto Roth [Abstract] AIS3 or [aiskju:b] is a contemporary response to the ancient idea of a music of the spheres connecting cosmology and music in novel way. The sound laboratory with its 444 illuminated spherical speakers, which is presented in late Summer 2019 at the Ludwig Forum für Internationale Kunst in Aachen, adopts the form and arrangement of the weirdest telescope in the world: IceCube consists of 5160 light sensors that are frozen deep into the Antarctic ice shield at the South Pole, covering a full cubic kilometre of ice. The sensors register the tiny flashes of light that occur in the rare case of an interaction of a very special type of elementary particle: a neutrino. In our joint presentation we show the way how these ”ghost particles” made their way from astro-particle physics to become the agents of an unique fundamental experiment in psychoacoustics, turning space into a sound generator in which the visitor becomes immersed. www.imachination.net/ais3

Session chair(s):

T. Dau, Technical University of Denmark

The ear at the age of IoT Jeremie Voix [Abstract] In this age of the Internet of Things (IoT), wearables are now everywhere, sometimes even in your earcanal. The research team from the NSERC-EERS Industrial Research Chair in In-Ear Technologies (CRITIAS) has been actively developing various in-ear technologies designed to complement the human ear, from ”smart” hearing protection against industrial noises, to advanced inter-individual communication systems, to hearing health monitoring devices using otoacoustic emission (OAE), to in-ear EEG Brain Computer Interface (BCI). More fundamental research has also been conducted, particularly on the micro-harvesting of electrical power from inside the earcanal to power future auditory wearables. The lastest research developments will be presented in this keynote presentation, together with some of the recent developments achieved through the CRITIAS industrial research chair in in-ear technologies or other research groups.

Session chair(s):

M. Ochmann, Beuth Hochschule für Technik Berlin

Thermoacoustic instabilities – physical mechanisms and mathematical modelling Maria Heckl [Abstract] If a flame is put into an acoustic resonator, an escalating interaction between the flame’s heat release and the acoustic field can occur, giving rise to intense pressure oscillations. This phenomenon is termed ”thermoacoustic instability”. It occurs in combustion systems that have a continuously burning flame, such as gas turbines, jet aeroengines, boiler and heating systems, furnaces and rockets. Thermoacoustic instabilities are a serious problem because they can lead to excessive structural vibrations, fatigue, and even catastrophic hardware damage. The escalating flame-sound interaction occurs in tandem with other physical mechanisms, leading to a complex web of interactions, most of which are nonlinear. They include flame-vortex interactions, flame response to fluctuations in fuel concentration, entropy waves, flame-structure interactions, and others. The development of low-pollution combustion systems is very important for our environment. However, such combustion systems are particularly prone to thermoacoustic instabilities. Progress with developing combustion systems that are immune to these is hampered by insufficient physical insight. Efforts to gain further insight are going on by researchers worldwide using experimental, analytical and numerical tools. This talk aims to give an overview of the key physical mechanisms involved in thermoacoustic instabilities and will pay particular attention to mathematical modelling approaches.

Organization:

J. Cheer, Institute of Sound and Vibration Research C. Shi, UESTC, Chengdu, China

	Effectiveness of ANC Partition with Film Speaker Yusuke Makiyama, Shun Hirose, Kohei Oto, Yusuke Komoto, Yoshinobu Kajikawa [Abstract] In this paper, we design a ANC partition with film speaker and demonstrate the effectiveness through experimental results. In a cubicle of public areas, like a reception desk inside bank, surrounding sound including conversation and environment noise avoids smooth conversation and the privacy may not be protected due to leaking to the adjacent cubicles. In such a case, the partition should have not only passive sound control function but also active sound control function. However, the ordinary loudspeaker (electrodynamic loudspeaker) may not be implemented into the partition effectively and may not control enough region in the cubicle. In this paper, we implement a film speaker into the partition and realize active noise control (ANC) system. The proposed ANC partition can achieve more than 10 dB noise reduction and wider zone of quiet compared with the ANC partition using the ordinary loudspeaker.
	Feedforward Control of Fan Noise in Ducts using Multichannel Order-reduced Inverse Filters Mingsian R. Bai, Shin-Cheng Huang [Abstract] Single-channel feedforward control is a commonly used approach for active noise control (ANC). In this paper, a time- domain underdetermined multichannel inverse filtering technique is proposed for the feedforward active control (ANC) of duct noise. In the commonly used filtered-x least-mean- square (FXLMS) algorithm, the feedforward control problem is formulated as an overdetermined inverse filtering problem which generally results in non-zero residual noise. By introducing multiple secondary sources, the problem can be reformulated into an underdetermined system, which admits infinite number of exact solutions with zero residual noise. Linearly constrained minimum variance (LCMV) method is employed in the controller design. However, as a major shortcoming of the time-domain approach, the length of the filters tends to be too long to admit digital signal processor (DSP) implementation. To address the problem, the least absolute shrinkage and selection operator (LASSO) algorithm, is exploited to effectively reduce the controller orders. A two-channel system is implemented to suppress the fan noise of an air-cleaner. Simulation and experiment results have demonstrated that the proposed approach has achieved significantly higher noise reduction than the conventional FXLMS algorithm.
	Integrated Simulation of Active Noise Cancellation using a Computational Fluid Dynamics Approach Arun Cherkkil, Vinod Narayanan, Nithin George [Abstract] An attempt has been made in this paper to simulate active noise control (ANC) in a three dimensional space using computational fluid dynamics, pressure acoustics and solid mechanics. We have considered a rectangular box, with primary and secondary sources at the two opposite sides of the box. A model of ANC has been developed, which considers a cardboard surface and air as the fluid. A control algorithm, which interacts with the fluid dynamic model has also been developed, which tries to achieve active noise control. The proposed approach can be used as an initial feasibility study to predict acoustic pressure field inside the control space.
	Limits of Noise Control over Space Thushara Abhayapala, Jihui Zhang, Prasanga Samarasinghe, Wen Zhang [Abstract] This paper investigates the best possible performance of noise cancellation over a region of space for a fixed set of secondary sources within a given acoustic environment. We first formulate the spatial active noise control (ANC) problem in a 3-D acoustic enclosure (room). Secondly, we derive a wave-domain least square method by matching the secondary sound field to the primary sound field in wave domain. Thirdly, we propose a subspace method by matching the secondary field coefficients to the projection of primary sound field in the subspace. Simulation results compare between the wave-domain least square method and the subspace method, in terms of energy of the loudspeaker driving signals, noise reduction inside the region, and residual noise field outside the region. The subspace matching method provides the limits of the ANC system for the given constraints and the acoustic characteristics of the room.
	Formation of local quiet zones using the length-limited parametric array loudspeaker Yue Wang, Ruicong Li, Chuang Shi, Youxin Li [Abstract] This paper presents the numerical study of forming local quiet zones using the length-limited parametric array loudspeaker (PAL). The PAL is a directional sound device using the nonlinear acoustic effect to generate an audio beam in air. The length-limited PAL targets to control the length of the audio beam by using two different ultrasonic carriers. The sound propagation curve and radiation pattern of the length-limited PAL are theoretically distinct from those of the conventional loudspeaker. Therefore, the shape of the local quiet zones formed by the length-limited PAL is of sufficient interest before any practical deployment. For comparison, the local quiet zones formed by the conventional loudspeaker and the normal PAL are also obtained in the numerical simulation.
	Robust stability and performance of local active control systems using virtual sensing Stephen Elliott, C K Lai, Thibault Vergez, Jordan Cheer [Abstract] Local active noise control systems often use virtual sensing techniques, to estimate the pressure to be controlled at the listener’s ears from an array of remote monitoring microphones for example. The virtual sensing process inevitably makes assumptions about the acoustic environment, although the environment is generally subject to uncertainties under different operating conditions. This paper considers the performance and stability of the remote microphone technique, as an example of a virtual sensing method, in a vehicle when subject to uncertainties due to changes in the internal acoustics and road conditions. It is shown, using data from models and experiments, that both the robust stability and the robust performance are improved by using regularisation in the design of the filter used to estimate the pressure at the listener’s ears from the outputs of the array of monitoring microphones.
	Tonal active control of the power scattered by locally-reacting spheres using a small number of radiators near the surface Mihai Orita, Stephen Elliott, Jordan Cheer [Abstract] A theoretical investigation is performed where spherical harmonic series expansions are used to describe the primary and secondary pressure fields in the problem of actively controlling sound scattering around spheres. The study focuses on the scenario of point-monopole radiators in the vicinity of a sphere with a uniform, locally-reacting and real-valued surface impedance. A control method based on minimizing the contributions of the spherical harmonic components to the scattered sound power is proposed. Under the established circumstances, any radiated sound power is found to be dominated by a few spherical harmonic terms of low degree in the region of large relative wavelength. As this wavelength decreases, numerous terms of higher degree become significant, thus increasingly more sources are required to achieve any suppression. A study case is performed where one point-monopole is placed at infinite distance to form an incident plane-wave. The scattering due to this is controlled with up to four point-monopoles on or near the surface of the sphere. Progressively better attenuation is achieved at large relative wavelength when increasing the number of control sources and the behavior can be described using asymptotes of the spherical harmonic coefficients.
	Synthesis of the Ineter by Direct Acceleration Feedback Neven Alujević, Ivan Ćatipović, Marko Jokić, Hinko Wolf [Abstract] Physical realisations of the inerter are often such that they must be either large scale, i.e. rack and pinion inerters, or they inherently include additional elements in parallel or in series with the inerter, i.e. shunted electromechanical transducers or active force feedback inerter realisations, or alternatively they bring along a large parasitic damping, i.e. fluid-based inerters. In this study inerter is realised by feeding back the subtracted outputs of two accelerometers attached to a reactive force actuator terminals. Although in theory such feedback loop is unconditionally stable due to the collocated sensor-actuator arrangement, in practice it may not exhibit good stability properties due to the lack of duality between the sensors and the actuator in conjunction with the internal dynamics of the transducers. Therefore in this paper dynamics of seismic accelerometer sensors and an electrodynamic actuator are fully incorporated into a theoretical model of such an inerter. The inerter model is coupled into a two degree of freedom mechanical system in order to study the stability of the feedback loop and the maximum possible synthesisable inertance. The results indicate that it is crucial to have a highly damped accelerometer resonance in order to achieve good stability and large synthesised inertance.

Organization:

S. Herold, Fraunhofer LBF E. Rustighi, University of Southampton

	Theoretical and Experimental Analysis of a Skyhook Damper for Active Control of Sound Transmission Neven Alujević, Steven Claes, Martina Šimag, Paul Sas [Abstract] In this paper active control of sound transmission through a double panel is considered. The active control system consists of one accelerometer mounted on the radiating panel whose time-integrated and amplified output is used as the error signal to drive a voice-coil actuator that reacts between the source and the radiating panel. Effects of one such active damper in the centre of the double panel system are considered, although, in principle, it could be repeated in the x-y plane and then used in the centralised, distributed or a decentralised manner. Theoretical and an experimental results are presented and compared. In the theoretical part, the mobility matrix model is used to investigate the stability of the feedback loop and to calculate the sound transmission ratio with and without control. It is shown that large reductions of the sound transmission ratio are possible, however only if the natural frequency of the fundamental source panel mode is higher than the natural frequency of the fundamental radiating panel mode. This is also discussed and shown analytically on a reduced order model. The experimental part of the study is carried out on the prototype active panel system and the measured result fully corroborate theoretical predictions.
	Hybrid Mass Damper Using Electromagnetic Resonator: Application to a Helicopter Simon Chesne, Guillaume Inquieté, Paul Cranga [Abstract] This work aims to illustrate how to increase the efficiency of an initial passive damped electromagnetic resonator by controlling the mass response actively. In aerospace and aeronautical engineering a fail-safe design for active systems is usually required. Hybrid Mass Dampers (HMD) tend to realize these objectives. Design rules and limitations of the so- called Hybrid Mass Damper (HMD) are discussed and analyzed. The main idea is to modify a DVA (Dynamic Vibration Absorber) to use it as an AMD (Active Mass Damper). Objectives are to take the best of the two technologies and increasing the performance of the passive device and/or decreasing the control effort of the actuator in comparison with a fully active system. A specific electromagnetic resonator is then designed and an appropriate control law is proposed. The methodology is illustrated and validated with an application to a helicopter. Theoretical and numerical results are completed by experimental validation.
	Active Engine Mount Stefan Loheide [Abstract] Conventional consumption and weight reduction measures aimed at an optimization of modern engine suspension systems, usually lead to an increase in vibrations acting on the powertrain suspension mounts. By application of active engine mounts these vibrations which are limited to special engine orders are decoupled from the subframe. The design and operating principle of the active engine mount are explained. Particular focus is on the selection criteria for the actuator as well as the interaction and coordination with the characteristic mount parameters. Another focal point of this representation is the description of the control system including the necessary boundary conditions. The operating principle and effect of active engine mounts in the complete system is demonstrated by taking results obtained from measurements in the vehicle into account.
	Simulation-Based Multi-Objective Optimization of a Fuzzy Controller for Semi-Active Suspension Louis Balzer, Valentin Mees, Jonathan Millitzer, Giovanni Lapiccirella [Abstract] This paper presents a simulation-based design process for a fuzzy-controlled semi-active suspension system applied to a real-time vehicle simulation. Firstly, a control architecture for the fuzzy-logic controller is defined. Secondly, different Genetic Algorithms (GAs) are configured for the optimization of the control parameters. A multi-objective optimization is performed in order to simultaneously improve safety and comfort of the driving vehicle, defined using two independent cost functions. The contact forces at the wheels estimate the safety and the vibration in the car body estimates the comfort. The performance of the controller is compared for each set of parameters obtained by the different GAs adopted. A reduced order real-time simulation environment has been set for a holistic vehicle simulation, which includes a Finite Element Model (FEM) of the chassis, non-linear suspensions, multi-body physics and the designed digital controller. Finally, the real-time environment is integrated into an overall optimization process and is used for fitness evaluation. The semi-active system under test allows to shift the Pareto front beyond the limit of passive systems, achieving simultaneously better safety and comfort. The different GAs come up with various near-optimal solutions, which are compared using their Pareto fronts.
	Design and applications of lean active resonator silencer cassettes Jens Rohlfing, Karlheinz Bay, Peter Brandstätt [Abstract] Active resonator silencer cassettes (ASCs) are composed of a housed loudspeaker coupled with a microphone sensor via an analog electronic control loop. ASC-units are very efficient in attenuating low-frequency noise, especially in applications with tight constraints on the available installation space. At the Fraunhofer IBP the design of ASC-units, ASC-arrays and their practical application has been investigated for more than two decades. This paper gives a brief overview on the characteristics, the performance and some practical applications of classical ASC-units and arrays. Classical ASC-units have dimensions in the range of decimetres. More resent work on ASCs at the Fraunhofer IBP has been focused on the design of much leaner units with dimensions in the range of centimetres. Scaling the dimensions of the ASC-units down by an order of magnitude widens the range of possible applications, but also comes with a number of design challenges. In this paper, these design challenges and the characteristics of the lean ASC-units and arrays are discussed. The paper also presents results of studies where the lean ASC-units are used to build up a compact active noise control system for a partly opened sliding window, and a compact active silencer system for a domestic ventilation duct.
	A boundary virtual sound barrier system for sound radiation through openings with double-layer secondary sources and error microphones Shuping Wang, Xiaojun Qiu, Jiancheng Tao [Abstract] Apertures, like vents of air-conditioning systems and slits of partially opened windows or doors are very common in the buildings, but the transmission loss performances are strongly influenced by the presence of these apertures. This paper proposes an analytical method to calculate the sound radiation from inside an enclosure through an aperture of finite thickness to the free field based on the modal superposition method and Kirchhoff-Helmholtz equation. The active noise control technique is then applied on the aperture to reduce noise transmitted through it to the outside. Different configurations of secondary sources and error microphones are investigated. It is found that sound radiation can be effectively reduced with an active noise system at the edge of the aperture, and the upper limit frequencies of effective control are discussed.

Organization:

H. Lissek, Signal Processing Laboratory, École polytechnique fédérale de Lausanne, Switzerland S.A. Cummer, Electrical and Computer Engineering Department, Duke University, Durham E. Ballestero, London South Bank University, The Acoustics Group

	Active Poro-elastic Acoustic Meta Materials Christopher Fuller [Abstract] Previous work has demonstrated the potential of poro-elastic acoustic meta materials for providing increased sound absorption at low frequencies[1]. This class of acoustic meta materials consist of a foam poro-elastic matrix supporting periodic arrays of microporous polymer sheets or small spherical masses. The increase in sound absorption and/or transmission loss occurs either by resonant wave scattering between the microporous sheets or the resonant dynamic motion of the embedded masses. In this paper we report preliminary work on extending the poro -elastic acoustic meta material to including active elements. The intention is to increase the bandwidth of the AMM attenuation, which is usually narrow, due it being based upon resonance conditions. In the active poro-elastic meta material some of the embedded masses are replaced with small inertial active vibrators with the same weight as the static mass. The control approach was a standard digital feedforward approach. Tests were carried out with narrowband and broadband plane wave signals in a standing wave tube. The results indicate there is potential in using active elements to broaden the frequency response of poro-elastic acoustic meta materials. 1. Fuller, C.R and Mathur, G.P., ”Poro-elastic Acoustic Meta Materials”, Proceedings of Inter Noise 2017, Hong Kong, August, 2017.
	Analysis and Optimisation of an Active Noise Control System as a Potential Acoustic Metamaterial Building Block Joe Tan, Jordan Cheer, Stephen Daley [Abstract] Active Noise Control (ANC) systems have been employed in a variety of applications to provide high performance noise control at low frequencies, within a lightweight and compact package. Recently, acoustic metamaterials (AMM) have been proposed and demonstrated as an alternative approach to achieving high levels of noise control. AMM are engineered structures that consist of an array of subwavelength unit cells, which exhibit behaviour not seen in conventional materials. For example, passive resonators have been designed to achieve negative effective material properties and manipulate wave propagation. These AMMs, however, typically have narrow band gaps, where wave propagation is forbidden. When acting as an AMM, ANC systems have been shown to improve performance, tuneability and adaptability, but physical insights have generally been limited. Therefore, this paper investigates the effects of traditional ANC on the effective material properties and shows physical insight through an analysis of the wave propagation within a one- dimensional duct system. The active unit cell has then been optimised to directly minimise the effective material properties and a corresponding physical analysis has been presented.
	Design of Acoustic Partitions with Thin Plate-like Acoustic Metamaterials Felix Langfeldt, Wolfgang Gleine [Abstract] Acoustic metamaterials have emerged as new means for sound control with extraordinary properties, such as negative effective density and/or bulk modulus. Amongst the variety of different realizations of acoustic metamaterials, thin plate-like metamaterials (e.g. membrane-type acoustic metamaterials or inhomogeneous plates) have a high potential for improving the sound transmission loss of conventional partitions, especially in the challenging low-frequency regime. These types of metamaterials have been previously shown to achieve very high sound transmission loss values which can exceed the corresponding mass-law values considerably. However, further investigations have shown that when these metamaterials are integrated into a double wall partition, the beneficial effect of the metamaterials can be greatly diminished if an improper design is chosen. This contribution aims at providing a more detailed understanding of the important parameters in the design of acoustic partitions with thin plate-like acoustic metamaterials. The metamaterial is represented by a simple black-box model in order to reduce the parameter space. Analytical calculations of combinations of the metamaterial with single and double walls are used to identify the most relevant design parameters and possible limitations in the acoustic performance. General design guidelines for acoustic partitions are derived from these results and applied to laboratory test samples.
	Non-linear Metamaterial Structures: Array of Particle Dampers Sifa Gul Demiryurek, Anton Krynkin, Jem Rongong [Abstract] Passive dampers to treat the excessive structural vibration has long been researched and used in industry. Multiple particles placed in a container can be used to dissipate excessive vibration through the inelastic collisions between particles and cavity of the damper has been shown. However, their application is usually limited to treating certain modes of structural vibration and their performance is highly dependent on location. Another limitation of the particle dampers is nonlinear character caused by discontinuity and randomness of the collisions and velocity of the particles. In this paper, it is proposed to modify a particle damper into a metamaterial type structure in order to expand the applicability range. Metamaterials are known to exhibit subwavelength performance offering superior vibro-acoustic properties over a wide range of frequencies. To maintain metamaterial properties, the casing of the particle damper is designed to resonate near selected modal frequencies. The Bloch-Floquet theory is applied in studying the doubly and singly periodic arrangement of the resonating damper shells with and without particles. Finally, the nonlinear effects observed in the metamaterial structure made of particle dampers are modelled numerically to predict their vibro-acoustic effects in finite structures. The theoretical, numerical predictions are compared with the experimental results.

Organization:

M. Pawelczyk, Institute of Automatic Control, Silesian University of Technology, Gliwice, Poland Y. Kajikawa, Kansai University

	Comparisons of Two Virtual Sensing Methods for Broadband Noise Reo Maeda, Yoshinobu Kajikawa [Abstract] Conventional Active Noise Control(ANC) reduces the target noise around the error microphone. Therefore, the noise reduction may not be enough at the desired position unless the error microphone is located there. To solve this problem, two virtual sensing techniques have been already proposed, which can obtain noise reduction at the desired position. One is called the remote microphone (RM) method; it’s valid especially for reducing the narrow band noise. However, there are two main issues in the RM method. When estimating the transfer function between microphones, some dips are generated at particular frequencies. Furthermore, the transfer functions cannot be estimated on some directions of noise arrival. On the other hand, the other method, which is called the auxiliary filter based virtual sensing (AF- VS), is more effective on reducing the broadband noise from all directions. In this paper, we compare the noise reduction performances of the two methods for broadband noise to clarify the effectiveness through simulations in MATLAB. As a result, both methods can shift the maximum noise reduction to the desired position, respectively. Comparing the two methods, AF- VS method can reduce the target noise better than RM one at the desired position on various arrangement.
	Virtual Sensing Technique for a Multi-Reference and Multi-Error Active Noise Control System Rong Xie, Chuang Shi, Huiyong Li [Abstract] This paper investigates the virtual sensing technique for the multi-reference and multi-error active noise control (ANC) system. Our previous study has validated the virtual sensing technique for a multi-channel feedforward ANC system where only one reference signal was available. In this paper, two reference signals are considered so that the numbers of control filters and auxiliary filters are doubled accordingly. A comparison is carried out between the case when 4 physical error microphones are placed to form 8 virtual error microphones and the other case when 8 physical error microphones are used to form 4 virtual error microphones. The limitation of the virtual sensing technique is thus concluded.
	A Method for Detecting Convergence Completion of Adaptive Filter Cancelling Feedback Path Appeared in Active Noise Control Systems Kensaku Fujii, Mitsuji Muneyasu [Abstract] We present a method for detecting the convergence completion of the feedback control filter working as an adaptive filter. Generally, the feedback path is manually estimated before starting active noise control. This estimation, however, is desired to be automatically completed. We first introduce a step size control method capable of completing the estimation with a prescribed error even where the power of disturbance fluctuates. Then, we propose a method for detecting the convergence completion of the adaptive filter. In practical use, since the impulse response of the feedback path and its power gain are unknown, we cannot directly monitor the difference between the impulse response and the coefficients of the adaptive filter. For solving this problem, we apply two adaptive filters with two different step sizes to estimating the feedback path. Actually, we have already presented a recursive filter expression illustrating the structure of the normalized least mean square algorithm. According to the expression, the square error between the coefficients of the two adaptive filters converges on the prescribed error multiplied by the power of the difference between the two step sizes. We finally verify using computer simulations that the proposed method can successfully detect the convergence.
	Practical consideration and implementation for avoiding Saturation of large amplitude Active Noise Control Shi Dongyuan, Woon-Seng Gan, Bhan Lam, Shulin Wen [Abstract] The saturation distortion of the power amplifier is a common and intractable issue for most audio devices, let alone for the active noise control (ANC) system. The effect of amplitude and phase distortion, caused by the saturated audio amplifier, on the adaptive algorithm in ANC is elaborated in this paper. To overcome this issue, the previously proposed the two-gradient FxLMS algorithm is applied, which efficiently suppresses the amplifier saturation and has as the same computational complexity as the FxLMS algorithm. The performance of the 2GD-FxLMS algorithm is experimentally validated for the first time in the control of both periodic and broadband signals.
	Empirical Study of Decentralized Multi-Channel Active Noise Control Based on the Genetic Algorithm Guoqiang Zhang, Jiancheng Tao, Xiaojun Qiu [Abstract] In an active noise control (ANC) system, computational complexity is one major concern when designing practical control algorithms. One approach to reducing computational complexity is to apply a decentralized control scheme rather than the centralized scheme. A decentralized scheme attempts to control a number of ANC subsystems independently, where for simplicity, one subsystem consists of one loudspeaker and one error microphone. Our recent published article has shown theoretically that decentralized two-channel ANC can achieve the same noise reduction performance as the centralized controller with guaranteed convergence in the frequency domain. In this work, we attempt to extend the results from two-channel case to N (N>1) channel case. The challenge sits in finding N complex numbers that could properly shape the eigenvalues of an N * N matrix for each frequency bin towards guaranteed convergence. Due to the problem complexity, we conduct empirical study by using the genetic algorithm (GA). Simulated results on the channel numbers of 2, 4, 6, and 12 demonstrate that the resulting decentralized ANC controller is also able to achieve the same noise reduction performance as the centralized controller.
	Design and Construction of Loudspeakers with Low-Bl Drivers for Low-frequency Active Noise Control Applications Marios Giouvanakis, Konstantinos Kasidakis, Christos Sevastiadis, George Papanikolaou [Abstract] The concept of the so-called low-Bl drivers is exploited, through the design and construction of small vented loudspeakers. Their high compliance and low-frequency resonance provide high- efficiency response in a limited frequency region and allow the implementation of small cabinet volumes. Their purpose is to be the elementary building components of a compound source for low-frequency noise control applications. The combination of such sources can lead to dipole or multipole set-ups, in which the radiation directivity pattern in open spaces and modal coupling in closed spaces can be controlled via the parameters of the individual driving signals. The construction constraints of the cabinet are accounted for, considering its design optimization. A low-volume cabinet with ports tuned close to the driver’s resonance frequency is constructed to extend the loudspeaker’s output at low frequencies. The impact of the position, the dimension and the number of the enclosure’s tubes to the loudspeaker’s frequency response are examined through measurements. The design analysis and experiments show the direction to their proper construction. The implementation of such loudspeakers for low-frequency noise control applications is advantageous in small rooms, where the available space is of great concern.

Organization:

E. Fernandez-Grande, Technical University of Denmark N. Xiang, Arch. Acoust. RPI, USA

	Acquisition of Bi-Directional Reflectance Functions by Nearfield Acoustical Holography – a preliminary study Jonathan Hargreaves [Abstract] It is well known that material absorption and scattering is dependent on incidence and observation angle. Despite this the corresponding standardised coefficients, which are used to represent these mechanisms within computational acoustic models, aggregate all such dependency into single random- incidence parameters. This limits the accuracy that can be achieved with computational acoustic models - even if these algorithms were to capture the wave physics perfectly, which they often do not, the results would not match physical reality because the input data is too low resolution. Bi-Directional Reflectance Functions are an established way of describing boundary absorption and scattering in computer graphics that have been suggested for use in acoustics. To date, a number of algorithms have been published that use these in simulation, but no measurement methods are available to acquire them. There is also ambiguity over some aspects of their definition e.g. whether finite panel size is included as a scattering mechanism. This paper adopts a definition suitable for high-frequency Boundary Element Method algorithms that use oscillatory basis functions to capture wave directions. It then proposes an acquisition method based on double-layer Near-Field Acoustical Holography and assesses it accuracy using 2D simulations.
	Improving in-situ sound absorption measurements using sparse multichannel blind deconvolution Bruno Masiero, Stelamaris Bertoli, Alvaro Pais [Abstract] The sound absorption of materials is traditionally measured in laboratory condition with one of two methods: random incidence in a reverberant chamber (ISO 354) or normal incidence in an impedance tube (ISO 10534). Nevertheless, there are some materials that cannot be measured in the lab, e.g., road surfaces, which are recommended to be measured in-situ with the use of a single microphone (ISO 13472). The latter method is based on time-windowing the measured impulse response to compare the incident and reflected wave components. Depending on the measurement setup, the size of the window may result in degraded measurement quality, specially at low frequencies. To alleviate this effect we use a sparse multichannel blind deconvolution approach. This methods blindly estimates the response of the sound source, resulting in a cleaner (in this case, more sparse) measurement that can than be used to estimate sound absorption. We compare the results of the proposed method with the results obtained with the single microphone method for a wall covered with PET-wool. The sound absorption coefficient was analyzed in the octave bands between 250Hz and 4000Hz.
	An application of multi-scale directional dictionaries to RIR interpolation Elias Zea [Abstract] The spatio-temporal sparsity of room impulse responses (RIRs) degrades in the late part due to the stronger wave interference compared with the early part. Meanwhile, such an interference decays in amplitude as time progresses due to absorption in the room, resulting in a decreasing dynamic range of the measurement. Together these aspects pose challenging conditions for compressive sensing applications, such as the interpolation of RIRs measured at sparse microphone positions. In the search of sparse transformation spaces, this paper examines the application of a multi-scale directional dictionary (known as shearlets) to interpolate RIR measurements. These redundant dictionaries consist of multiple curved elementary functions, which offer a decomposition of the acoustic wavefronts into various wavelengths, propagation directions, and times of arrival. Results reported in this paper demonstrate the potential these dictionaries have to interpolate RIRs in both convex and nonconvex rooms, motivating further examination under experimental conditions and in broader frequency ranges.
	Higher-order processing of spatial impulse responses Leo McCormack, Archontis Politis, Oliver Scheuregger, Ville Pulkki [Abstract] This article primarily concerns parametric methods for rendering multi-channel impulse responses for multi-channel loudspeaker playback of anechoic recordings. These methods analyse a set of spatial parameters, divide the response into components according to the assumed sound-field model, and render the components to the loudspeaker setup using methods informed by the analysis results. For a direct/diffuse model, the sound is divided into non-diffuse and diffuse components, which are reproduced over directional and surrounding reproduction methods, respectively. In many cases, the inputs are first divided into frequency bands for the analysis and reproduction. In this article, some acoustic conditions, which are especially critical of parametric methods, are identified, and the effect of time-frequency processing and diffuse-stream processing is investigated. In addition, a method that utilises higher-order spherical microphone components is described, which, according to the analysis results, overcomes the issues highlighted/emphasised by these problematic acoustic conditions. The method divides higher- order signals into multiple directionally biased first-order spherical harmonic signals, which are then analysed independently. The non- diffuse parts are reproduced individually for each sector, and the diffuse parts are encoded back into a higher-order signals, and subsequently reproduced using a matrixing method, followed by phase decorrelation.
	Deep Learning Applied to Dereverberation and Sound Event Classification in Reverberant Environments Mingsian R. Bai, Wen-Chuan Chen [Abstract] This paper investigates dereverberation and sound event detection techniques, with the aid of deep learning. The system consists of two units: a microphone array front- end and a deep learning back-end. The system is examined in the context of two important acoustic signal processing: dereverberation and sound event classification (SED). For the dereverberation problem, a neural network-based approach is compared with other state-of-the-art methods such as beamforming, the multi-channel inverse filtering (MINT), the multi-channel Wiener filter (MWF), and the variance-normalized delayed linear prediction (NDLP). For the SEC problem, two approaches are also compared. Approach 1 is based on the signals enhanced by the microphone array front-end, which serves as the input to the back-end deep neural network (DNN) classifier. The DNN is implemented by using the VGGNet®. Approach 2 is a direct approach that uses reverberant data directly to train the classifier without the front-end enhancement. The audio features are extracted as MFSC from the AudioSet from Google®. A room response simulator based on the image source method is employed to create reverberant signals for numerous RT60 conditions in the training phase. Perceptual evaluation of speech quality (PESQ) and F1-Score are adopted to assess the audio quality and classification performance.
	Binaural Dereverberation Based on Delayed Widely Linear Prediction in the Time Domain Xin Leng, Jingdong Chen, Jacob Benesty [Abstract] Sound spatial information benefits human listeners in reverberant environments. In this paper, we deal with the problem of binaural dereverberation, which reduces reverberation and meanwhile preserves the spatial information of the sound source at the binaural outputs. Specifically, a widely linear (WL) filtering framework is adopted where the multiple real microphone signals are merged into complex signals. By doing so, the desired binaural outputs are also converted into complex signals with one channel being its real part, and the other channel being its imaginary part. In such a framework, the complex late reverberation is modeled using the multichannel delayed WL prediction by fully taking advantage of the noncircularity of the complex signals. The maximum likelihood method is then used to estimate the optimal prediction filter with the speech signal of interest being modeled by a complex normal distribution. The relationship between the proposed method and the weighted prediction error (WPE) method is also discussed. Finally, simulation results are provided to justify the effectiveness of the developed method.
	Dereverberation Based on Deep Neural Networks with Directional Feature from Spherical Microphone Array Recordings Jeongmin Liu, Byeongho Jo, Jung-Woo Choi [Abstract] The dereverberation of a reverberant audio signal can be accomplished through the deconvolution of a room impulse response (RIR) from the reverberant signal. However, the room impulse response is unknown in most practical situations, which makes the dereverberation as a challenging problem. In this work, we tackle the dereverberation problem by incorporating a learning-based algorithm and directional input features measured by spherical microphones. The previous dereverberation studies utilizing the deep neural network (DNN) have shown that the performance strongly depends on RIRs used for the training data, because the trained network cannot generalize various room conditions. To build a more generalized model, we use directional features recorded by a spherical microphone array as the input to the DNN. Since the reverberation includes sound waves propagating to many directions, the use of directional cues can provide useful information on the room condition, thereby helping to construct a more generalized dereverberation model. When the DNN is trained by 500 speech utterances convolved with 20 RIRs and tested with 20 unseen RIRs, the use of directional features greatly improves short-time objective intelligibility (STOI) and perceptual evaluation of speech quality (PESQ) as compared to the same model trained by single channel features.
	Acousto-optic sensing – spatial reconstruction of the sound field enclosed in a room Samuel Arturo Verburg, Efren Fernandez-Grande [Abstract] Measuring the sound pressure field over medium and large volumes, such as rooms, can be cumbersome. Conventional sensing methods acquire the sound pressure (or other field quantity) at single a point. Therefore, to acquire the three-dimensional sound field in a room, a large number of measurement points must be distributed throughout. In this study, we propose a method to capture the sound field in a volume based on the acousto-optic effect. The acoustic pressure field is captured along beams of laser light. As the light (the sensing element of this method) travels through the sound field, it captures information of the sound pressure along its travel path. In this study, the spatial reconstruction of the sound field in a room from acousto-optic measurements is presented and compared with conventional point-wise measurements. The results show that the proposed acousto-optic sensing method is well-suited for the spatial characterization of sound fields over large volumes.
	A High Order Rigid Spherical Microphone Array Design Using MEMS Microphones Marco Berzborn, Michael Vorländer [Abstract] In recent years spherical microphone arrays have gained much prominence in the field of room acoustics. They provide a powerful framework for the analysis of the directional properties of sound fields in rooms. However, their angular resolution is limited by the number of microphones in the array, resulting in a poor performance for low order arrays when used in highly reverberant sound fields found in reverberation rooms. However, higher order microphone arrays require a large quantity of microphones potentially rendering the design economically unfeasible. We present a high order spherical microphone array design using consumer grade MEMS microphones combined with an equalization kernel implemented on an FPGA. A performance analysis of the prototype regarding the usable frequency range and the achievable angular resolution is given.
	Robust localization of early reflections in a room using semi real-valued EB-ESPRIT with three recurrence relations and Laplacian constraint Byeongho Jo, Jung-Woo Choi [Abstract] To localize early reflections in a reverberant room, the direction- of-arrival (DOA) estimation technique has been evolved in various ways. Among many techniques, the subspace- based localization techniques such as eigenbeam estimation of signal parameters via rotational invariance techniques (EB- ESPRIT) have been extensively researched for their ability to directly estimate DOA parameters from spherical microphone recordings. The EB-ESPRIT estimates DOAs by computing the directional parameters from a single recurrence relation of spherical harmonics. For estimating DOAs of early reflections, however, the EB-ESPRIT is inappropriate due to the insufficient number of detectable echoes and its vulnerability to noise. In this study, we attempt to overcome these issues by using three different recurrence relations of spherical harmonics and extra constraints. The proposed technique can compute the directional parameters more robustly by imposing Laplacian constraint derived from the Helmholtz wave equation. We show that constraint imposed on the joint eigenvalue decomposition (JEVD) algorithms can further enhance the accuracy of DOA estimation as compared to all EB-ESPRIT based techniques. In simulations, we analyze the number of detectable echoes and validate the robustness of the proposed technique in terms of the estimation error under different signal- to-noise ratios (SNRs) and direct-to-reverberant energy ratios (DRRs).
	Double Tetrahedral Intensity Probes for Reducing the Spatial Bias Error of Source Localization In-Jee Jung, Jeong-Guon Ih [Abstract] A compact tetrahedral probe used for measuring the three-dimensional acoustic intensity vector can be used for the source localization. Although such 3D intensimetry is advantageous in miniaturizing the sensing system size, it has not been popularly used due to its large spectral and spatial bias errors, which are additional to the bias errors at low and high frequency ranges. Compensation methods for spectral bias errors are recently proposed, but the spatial bias error, which is related to the probe orientation, is not easy to compensate. In this work, the idea is to adopt two probes together, thus effectively arranging the microphones to reduce the spatial sparseness of sensors and the irregularity in directivity. The number of microphones is minimized by sharing 1 or 3 microphones in the double module probes, while permitting the calculation of two intensity vectors. Two different types of probe configurations are used: twisted double probes, double tetrahedral probe symmetric to a face of the tetrahedron. A numerical simulation is conducted to compare the proposed probe systems with single probe. The result shows that the residual spatial bias error is less than 7.5∘ for 2.5<kd<4.1 range.
	Sound field reconstruction in a room from spatially distributed measurements Efren Fernandez-Grande [Abstract] The full characterization of the sound field in a room from experimental data is a challenging problem, as the direct measurement of the three-dimensional sound field in the volume requires considerable experimental effort. In this study, the sound field in a room at low frequencies is characterized using a representation in wavenumber domain that infers the modal structure of the sound field. The modal structure is used to synthesize frequency response functions, to characterize the spatio-temporal properties of the wave field. The proposed method makes it possible to infer frequency response functions between source-receiver pairs that have not been measured, exploiting the representation of the sound field as a modal superposition.

Organization:

M. Cobos, Universitat de Valencia D. Kolossa, Ruhr University Bochum

	Probabilistic Modeling for Learning-based Distance Estimation Andreas Brendel, Andy Regensky, Walter Kellermann [Abstract] Estimating the position of an acoustic source is an essential step for many signal processing applications. Hence, many approaches for acoustic source localization have been proposed in recent years, whereby most of them are based on the estimation of the direction of arrival of the acoustic source relative to an observing microphone array. In contrast, the estimation of the source-microphone distance is much less investigated. Especially for complex sound fields in acoustic enclosures, a promising approach is to apply machine learning approaches to learn the mapping of a distance-related feature to the corresponding source-microphone distance. Here, we focus on a recently proposed method for distance estimation based on the coherent-to-diffuse power ratio and Gaussian process regression. We investigate the influence of different probabilistic models reflected by the choice of kernel and mean functions of the Gaussian process employed for regression. The influence of these choices is quantified by experiments.
	Speaker Distance Estimation using Binaural Hearing Aids and Deep Neural Networks Mehdi Zohourian, Jakob Stinner, Rainer Martin [Abstract] In this paper, we address the problem of speaker distance estimation using binaural hearing aid microphones. The proposed approach is based on deep neural networks which aim to classify discrete distances at different levels of precision. We investigate two types of networks, namely, feedforward and convolutional neural networks which are trained on the binaurally estimated direct-to-reverberant energy ratio. The performance of the proposed algorithm is assessed on speech signals convolved with both synthetic and real binaural room impulse responses for several distances and in different acoustical scenarios. On the one hand, the proposed approach offers a reliable classification of coarse distances classes which is also robust against variations of the reverberation time. On the other hand, when a more accurate distance estimation is required, the proposed approach is robust only against small variations of the reverberation time.
	Binaural Direction-of-Arrival Estimation in Reverberant Environments Using the Direct-path Dominance test Hanan Beit-On, Boaz Rafaely [Abstract] Speaker localization in reverberant environments is an important and current challenge in audio signal processing with several recently methods proposing solutions to this challenge. One such family of methods is based on the direct-path dominance (DPD) test and was developed specifically for spherical arrays. In this paper, the DPD test based method is theoretically extended to arbitrary arrays and is specifically studied for a binaural array, showing good performance under diverse reverberation conditions. This extension of the DPD test based approach uses a spherical harmonics decomposition of the steering matrix to remove the frequency dependence of this matrix within frequency bands using focusing. This facilitates frequency smoothing which is necessary for decorrelating coherent reflections. The success of this focusing process may be affected by the frequency bandwidth as well as the array configuration. This paper presents an analysis of the performance of this focusing process and investigates its impact on the DPD test performance for binaural arrays. It is shown that the frequency bandwidth necessary for frequency smoothing may be sufficiently small such that the focusing stage may be skipped when applying the DPD test for binaural arrays. A simulation study that validates this result is presented.
	Data-driven Threshold Selection for Direct Path Dominance Test Orhun Olgun, Huseyin Hacihabiboglu [Abstract] Direction-of-arrival estimation methods, when used with recordings made in enclosures using rigid spherical microphone arrays (RSMAs), is negatively influenced by the reflections and reverberation in that enclosure. Direct path dominance (DPD) test was proposed as a pre-processing stage which can provide better DOA estimates by selecting the time-frequency bins with a single dominant sound source component thereby reducing the total computational cost. DPD test involves selecting bins for which the ratio of the two largest singular values of the local spatial correlation matrix is above a threshold. The selection of this threshold is typically carried out in an ad hoc manner, which hinders the generalisation of this approach, potentially increasing the total computational cost or reducing the accuracy of DOA estimation. We propose a DPD test threshold selection method based on a data-driven statistical model of singular value ratios. The model is based on the approximation of the singular value ratios as a generalised Pareto distribution and allows selecting bins based on their probability of occurrence. We demonstrate the application of this threshold selection method using numerical simulations, emulations using measured acoustic impulse responses and a real recording using an RSMA.
	Source Localization using a Spatial Kernel based Covariance Model and Supervised Complex Nonnegative Matrix Factorization Antonio Jesús Muñoz Montoro, Violeta Montiel-Zafra, Julio José Carabias-Orti, Juan Torre-Cruz, Francisco Jesús Canadas-Quesada, Pedro Vera-Candeas [Abstract] This paper presents an algorithm for source localization using a beamforming-inspired spatial covariance model (SCM) and complex non-negative matrix factorization (CNMF). The spatial properties are modeled as the weighted sum of spatial kernels which encode the phase differences between microphones for every possible source location in a grid. The actual localization for each individual source in the multichannel mixture is estimated using complex- valued non- negative matrix factorization (CNMF) where each source spectrogram is modeled using a dictionary of spectral patterns learned a priori from training material. Localization performance of the proposed system is evaluated using a multi-channel dataset with configurations (number of simultaneous sources, reverberation time, microphones spacing, source types and spatial locations of the sources). Finally, a comparison to other state- of-the-art localization methods is performed, showing competitive localization performance.

Organization:

M. Cobos, Universitat de Valencia D. Kolossa, Ruhr University Bochum

	Towards low cost acoustic cameras for the Internet of Things Jose Javier Lopez, Maximilian Becker, Carlos Hernandez [Abstract] In the context of Internet of Thing (IoT) accurate and low-cost sound source localization methods are of great interest, alone or combined with video images. In this work, we present an implementation of the Steered-Response Power Phase Transform (SPR-PHAT) method employing a small circular array of 10 cm consisting of 8 MEMS microphones all commanded by a small single-board computer (Rasperry Pi). The localization is performed both in azimuth and elevation in one side of the array. A camera emplaced in the center of the array can be also employed for multimodal localization. The evolution of the localization accuracy as more microphone pairs autocorrelacions are added in the estimation of likelihood is compared. An optimal number of autocorrelation pairs in the sense of accuracy/computational cost tradeoff is obtained.
	Inference in Nonlinear Dynamical Systems with dynamic Stream Weights for Audiovisual Speaker Tracking Christopher Schymura, Dorothea Kolossa [Abstract] Audiovisual speaker tracking is a prominent example of a multimodal signal processing framework with a large variety of technical applications. Whereas acoustic speaker tracking suffers from decreased performance under adverse acoustic conditions, visual sensors are not affected by purely acoustic disturbances. However, the optimal combination of acoustic and visual information remains a challenging task due to the highly dynamic nature of audiovisual scenes. Recently, a framework for audiovisual speaker localization and tracking has been proposed, which extended the notion of dynamical systems with adaptive stream weights. It essentially implements a weighting for acoustic and visual observations that can be changed at each time step and allows probabilistic inference based on the Gaussian filtering paradigm. This study presents a detailed analysis of the inference procedure and proposes an extension for nonlinear dynamical systems inspired by the unscented Kalman filter. Under the constraint that the process and observation noise terms follow a zero-mean Gaussian distribution, the inference framework presented in this study allows an efficient estimation of the speaker trajectory from audiovisual observations. A systematic experimental evaluation assesses the performance of the proposed method using recorded audiovisual scenes.
	Audiovisual active speaker localization and enhancement for multirotor micro aerial vehicles Daniele Salvati, Carlo Drioli, Andrea Gulli, Gian Luca Foresti, Federico Fontana, Giovanni Ferrin [Abstract] We address the problem of localizing a speaker and enhancing his voice using audio-visual sensors installed on a multirotor micro aerial vehicle (MAV). Acoustic-only localization and signal enhancement through beamforming techniques is especially challenging in this conditions, due to the nature and intensity of disturbances originated by the electrical engines and the propellers. We propose a solution in which an efficient beamforming-based algorithm for both localization and enhancement of the source is paired to a video-based human detection. The video processing front-end detects the human silhouettes and provides an estimation of direction of arrivals (DOAs) on the array. When the acoustic localization front-end detects a speech activity originating from one of the possible directions estimated by the visual components, the acoustic source localization is refined and the recorded signal is enhanced through acoustic beamforming. The proposed algorithm was tested on a MAV equipped with a compact uniform linear array (ULA) of four microphones. A set of scenes featuring two human subjects lying in the field of view and speaking one at a time is analyzed through this method. The experimental results conducted in stable hovering conditions are illustrated, and the localization and signal enhancing performances are analized.
	Study on Large Scale Projectile Impact Point Positioning Method Based on Trajectory Shock Wave Difeng Sun, Xubin Liang, Tianqing Zhao, Houlin Fang, Cheng Zhang, Hui Zheng, Liangyong Zhang, Fang Zhang, Deyu Sun, Yang Liu [Abstract] Abstract: With multiple nodes of single acoustic sensor layed around the target, the ballistic shock wave signal of the projectile generated before hitting the ground can be detected at different locations. The trajectory can be estimated by measuring the time difference between the signal of different nodes, and the impact point can be coordinated by crossing the trajectory with the target. In this article, the projectile impact point positioning model is parameterised and fomulated by nonlinear equations with six patameters, and it can be solved by means of nonlinear least square(NLS) method. By means of Cramer-Rao lower bound (CRLB) analysis, measurement error distribution by different incident angle and different impact location is calculated, and also, parameter of exterior ballistics model and wind vector is under consideration.

Organization:

S. Van De Par, University of Oldenburg S. Doclo, Dept. Medical Physics and Acoustics, University of Oldenburg

	Binding of speech syllables when segregation occurs Marion David, Mathieu Lavandier, Nicolas Grimault, Andrew, J. Oxenham [Abstract] Two previous studies [David et al., 2017, Hear. Res. 344, 235-243; David et al., 2017, J. Acoust. Soc. Am. 142(3), 1674-1685] have investigated the segregation of speech syllables made of a fricative consonant and a voiced vowel, referred to as CV tokens. The first study explored the segregation of such syllables based on fundamental frequency differences. In the second study, the authors assessed the segregation of CV tokens based on localization cues, especially the spectral cues in the median plane. In both studies they found that segregation can be observed based on F0 and on spectral cues. More interestingly, they found that the whole CV token stay grouped even when segregation occurs based on cues that affect only one part of the CV: F0 differences affect mostly the vowel part, whereas coloration in the median plane is effective mostly at high frequencies, which selectively affects the consonant part. The mechanisms that allow the CV to remain grouped under such circumstances remain unclear. The present manuscript reviews the results of these two studies and provides some suggestions as to how such binding might occur.
	Musical scene analysis of hearing-impaired and normal- hearing listeners: a melody and instrument matching task Kai Siedenburg, Saskia Röttges, Kirsten Wagener, Volker Hohmann [Abstract] It is well known that scene analysis is at the core of hearing- impaired (HI) listeners’ difficulties of understanding speech in noise. Surprisingly little is known about scene analysis of HI listeners in musical contexts. Here we tested 24 moderately HI listeners and 28 normal-hearing (NH) listeners in their ability to identify instruments and melodies in the presence of a musical accompaniment that acts as a masker (cello/piano dyads or spectrally matched noise). Target signals consisted of four-note melodies. In each trial, a signal-masker mixture was presented followed by two different versions of the signal without masker. Listeners judged which signal version was part of the mixture. Signal versions either differed in terms of timbre (flute vs. trumpet) or in terms of melody. Signal-to-masker thresholds (71% correct response rate) were measured by varying the signal presentation level in a 2down-1up procedure. The masker level was set to 65 dB SPL for NH listeners and to medium loudness for HI listeners. We observed drastically elevated thresholds for HI listeners (>+10dB) compared to NH listeners. Participants with musical training generally exhibited significantly lower thresholds. These results for the first time demonstrate drastic effects of hearing impairment on musical scene analysis.
	Comparative Study of Single-Channel Algorithms for Blind Reverberation Time Estimation Heinrich Löllmann, Andreas Brendel, Walter Kellermann [Abstract] Knowledge about the reverberation time (T60) is exploited by numerous algorithms, e.g., for automatic speech recognition, speech dereverberation, or as a feature for acoustic scene classification. This contribution provides a comprehensive survey about various state-of-the-art methods to estimate the reverberation time blindly from a noisy and reverberant speech signal and compares their estimation performance for different acoustic scenarios. The evaluation considers different acoustic conditions regarding the signal-to-noise ratio (SNR), direct-to-reverberation ratio (DRR) and T60, and uses databases with room impulse responses measured in different acoustic enclosures. The findings of the Acoustic Characterisation of Environments (ACE) Challenge are extended by taking into account algorithms and acoustic scenarios which are not considered in this benchmarking campaign, and by using a much larger set of reverberant speech signals.
	The effect of room acoustics on audio event classification Dimitra Emmanouilidou, Hannes Gamper [Abstract] The increasing availability of large-scale annotated databases, together with advances in data-driven learning and deep neural networks, have pushed the state of the art for computer-aided detection problems like audio scene analysis and event classification. However, the large variety of acoustic environments and their acoustic properties encountered in practice can pose a great challenge for such tasks and compromise the robustness of general-purpose classifiers when tested in unseen conditions or real-life applications. In this work we perform a quantitative analysis of the effect of room acoustics on general audio event detection scenarios. We study the impact of mismatches between training and testing conditions in terms of acoustical parameters, including the reverberation time (T60) and the direct-to-reverberant ratio (DRR), on audio classification accuracy and class separability. The results of this study may serve as guidance for practitioners to build more robust frameworks for audio event classification tasks.
	Deep network source localization and the influence of sensor geometry Jörn Anemüller, Hendrik Schoof [Abstract] Learning-based localization approaches cast the acoustic speaker localization problem as a machine learning task where a classifier is trained on example data of acoustic feature vectors in order to predict likelihood of speech presence as a spatio-temporal distribution. We investigate the impact that fundamental acoustic parameters of the auditory scene (e.g. SNR, acoustic scene complexity, sensor geometry) exert on the ability to faithfully extract spatio-temporal activity maps for concurrent speakers. Our results indicate that to some degree shortcomings in the acoustic conditions can be compensated by increased complexity in the applied classification techniques. To this end, we systematically investigate localization performance for a set of deep neural network localizers of varying complexity, and for six different sensor configurations in a bilateral hearing aid setup. Deep networks result in improved performance compared to linear localizers, and their performance benefits more from an increase in the number of sensor channels. In specific configurations, deep networks with a smaller number of microphones perform better than a linear baseline network with a larger number of microphones. Thus, location-specific information in source-interference scenarios appears to be encoded non-linearly in the soundfield, requiring non-linear approaches for optimal decoding.
	Virtual auditory scenes created by time reversal mirror technique Georgina Alejandra Lizaso, Jorge Petrosino [Abstract] The present paper describes the process of generating virtual auditory scenes exhibiting multiple virtual sources in different locations, which is accomplished through the application of the time reversal mirror (TRM) method. This technique, developed by Mathias Fink, can be used to focus an acoustic signal at a particular point in space. Time-reversing the transfer function between a TRM array and an acoustic source generates an acoustic spatio-temporal focus at said source’s original location. Thus, this time-reversed focus behaves as a ”virtual” source in the outbound direction with respect to the TRM. Provided that an acoustic impulse is previously registered by the TRM device, a ”virtual” audio source can be generated at the impulse’s location by convolving the TRM impulse response with an audio signal. Since the system is linear, it allows the addition of impulse responses belonging to different locations, which can be convolved with audio signals in order to shape the sound field of the auditory scene. The numerical simulations implemented to explore this method, located arbitrary audio signals in selected positions of an auditory scene. Their results were evaluated via comparison of the data from spatially localized sources against the virtual sources generated by the TRM technique.

Organization:

J. Ahrens, Chalmers University of Technology

	Audio Spotlight using Subdivided AM Sideband Wave Delivery from Separate Ultrasonic Array Speakers Kazuhiro Kondo, Shinichiro Osanai [Abstract] Parametric array speakers can deliver an audible sound beam by modulating sound using an ultrasonic carrier wave and playing this modulated signal from an ultrasonic transducer array at high levels. Audible sound is demodulated in a narrow beam due to the nonlinear aerial transfer characteristics. However, if sound can be delivered within a small spotlight, sound can be delivered only to the intended subject, and avoid leakage to others. This is partially possible by generating the carrier wave and the sideband signal of the modulated signal from separate arrays, resulting in the full demodulation of an audible sound at the intersection of the two beams. However, some audible noise was generated through demodulation of the sideband signal. The level of this noise is proportional to the relative frequency difference of the sideband signal. Thus, we divided the sideband signal into several sub- bands, limiting the frequency range of each sub- band, and generated them from separate arrays, intersecting only at the spotlight. By using 5 sub-bands, it was possible to limit the spotlight to an effective area of 0.2m by 0.2m, while the audible noise of the surrounding area was reduced to 20 dB lower levels.
	Evaluation of Thermal Runaway Control Based on Frequency Modulated Carrier Wave in Parametric Array Loudspeaker Kairi Mori, Masato Nakayama, Takanobu Nishiura [Abstract] A parametric array loudspeaker (PAL) can realize a sharper directivity by utilizing an ultrasound wave (carrier wave). The conventional PAL has a problem that ultrasonic transducers cause thermal runaway by strongly emitting an amplitude modulated (AM) wave through a long period of time. Therefore, we have proposed a thermal runaway control method with a frequency modulated carrier wave. In the proposed method, we generate the AM wave with the frequency modulated carrier wave that the frequency of the carrier wave changes with time. By changing the frequency of the carrier wave with time, we can minimize the effects of thermal runaway in ultrasonic transducers. In the paper, to evaluate the effects of thermal runaway, we perform evaluation experiments for temperature and the frequency response of the PAL with the conventional and the proposed methods. Specifically, we compare the change in the frequency response of the PAL before and after emitting the carrier wave of pure tone and the frequency modulated carrier wave for a long period of time, and the change in temperature of the PAL at that time. As a result, we confirmed the effectiveness of the proposed method.
	Three-dimensional Resonance Control Based on Spatial Wave Synthesis with Parametric Array Loudspeaker Shiori Sayama, Masato Nakayama, Takanobu Nishiura [Abstract] A parametric array loudspeaker (PAL) realizes a sharper directivity utilizing the demodulation of intense amplitude modulated (AM) wave. The demodulation is modeled as spatial sound sources generated with the demodulated audible sound in the acoustic beam. Spatial sound sources form resonance and anti-resonance positions in the acoustic beam because of the interference between spatial sound sources. Thus, the sound pressure level (SPL) of the PAL is reduced at the anti-resonance position. However, it suggests that resonance positions might be reformed by spatial wave synthesis with phased array processing. Therefore, we propose three-dimensional resonance control based on spatial wave synthesis with the PAL. In this paper, we employ the PAL consists of line-type PALs corresponding to column. The PAL can apply different delay filters to each column. The proposed method designs inverse filters with a multichannel adaptive algorithm as delay filters for line-type PALs under the constraint condition that the SPL is maximized by spatial wave synthesis at the target position. The proposed method can improve the SPL at the target position by applying the designed filters to each column. Finally, we confirmed the effectiveness of the proposed method through evaluation experiments.
	Demodulation Distance Control Based on Analytic Model between Film Gas-lens Depth and Demodulation Distance for Parametric Array Loudspeaker Toshihiro Fujii, Masato Nakayama, Takanobu Nishiura [Abstract] A parametric array loudspeaker (PAL) can transmit an audible sound to only a particular area by utilizing an ultrasound. We have previously proposed a demodulation distance control using a film gas-lens with the gas different to the air to control a maximum demodulation distance (MDD), which is a distance with maximum demodulation of the audible sound. In the previously proposed method, the MDD has a large numerical error because it employs a simple linear approximation. In this paper, we therefore propose a new MDD control based on an analytic model between film gas-lens depth and the MDD. To design the analytic model, we investigate the relationship between film gas-lens depth and the MDD through the preliminary experiments. Also, we design a regression curve of an exponential function into relationship between input film gas-lens depth and output MDD. Next, to calculate the relationship between input MDD and output film gas-lens depth, the analytic model is calculated as inverse function of the regression curve. The analytic model can accurately adjust the film gas-lens depth to control the MDD. Finally, we confirmed the effectiveness of the proposed method through the evaluation experiments.
	Development of Multi-way Parametric Array Loudspeaker Using Multiplexed Double Sideband Modulation Yuting Geng, Yusei Nakano, Masato Nakayama, Takanobu Nishiura [Abstract] Parametric array loudspeaker (PAL) can achieve a sharper directivity by utilizing ultrasonic wave than conventional dynamic loudspeakers. However, PAL has difficulty reproducing low-frequency sound because it utilizes demodulation of ultrasound in the air. We have proposed the multiplexed double sideband (M-DSB) modulation for PAL, which can enhance sound pressure level (SPL) by utilizing harmonic distortion in demodulation. However, in lower frequency, sound quality by PAL with M-DSB is still insufficient compared with one by dynamic loudspeakers. Therefore, in this paper, we develop a new multi-way PAL with combining software and hardware approaches, which can achieve both sharper directivity and powerful bass sound. Multi-way PAL consists of a tweeter PAL and woofer PALs, and each part is an aggregate of multiple ultrasonic transducers. Tweeter PAL employs single sideband (SSB) modulation, which gives low-distortion but low-SPL. Woofer PAL employs M-DSB modulation, which gives high-SPL but high-distortion. In order to achieve a flatter frequency response, we conducted preliminary experiments to determine both a suitable number of ultrasonic transducers and an optimized arrangement for tweeter PAL and woofer PALs. Finally, we confirmed the effectiveness of the proposed method through evaluation experiments.
	Design of a directive source using the radiation mode method and a sound zone algorithm Manuel Melon, Maryna Sanalatii, Philippe Herzog, Régine Guillermin, Nicolas Poulain, Jean Christophe Le Roux [Abstract] A common mean to achieve narrow directive patterns consists in using loudspeaker arrays. Generally, the position of the sources in the array is arbitrarily fixed, for instance by using a regular spacing. The filters for each speaker are then calculated to get the desired acoustic field. In some cases, the predefined speaker positions might not be the optimal ones. Moreover, the diffraction on the source body may not be negligible for medium range applications. In this paper, we propose a two-step approach to overcome these problems. First, the radiation modes of the enclosure are computed. They are used in an inverse scheme, to find the most efficient vibration patterns over the enclosure, able to generate a target field at some chosen frequencies. These vibration patterns are then used to define the positions and sizes of actual loudspeakers on the enclosure. In a second step, a sound zone algorithm is used to calculate the signals driving the speakers. This yields a very narrow directivity pattern with reduced side lobe amplitudes. The proposed method is tested both numerically and experimentally and results are reported. A discussion about the pros and cons of the proposed approach is then provided.
	Models of spherical and rectangular cuboid loudspeaker arrays Franz Zotter, Hannes Pomberger [Abstract] Compact spherical/cubical loudspeaker arrays are used as beamformers to radiate focused sound with adjustable direction. A common enclosure housing all the transducers is easier to manufacture but acoustically couples their motion, making their independent control more challenging Nevertheless, recetly analytic IIR filters to controll cubical arrays were proposed, which motivates that such could be defined for more general arrays in the future. As a first step, this contribution describes the analytic model of the interior crosstalk in (i) rectangular cuboid loudspeaker arrays using piston transducers, and in (ii) sperical loudspeaker arrays using spherical cap transducers. Both models are based on interior modes and their accurate integrals to evaluate both the contribution of the transducers’ velocity boundary conditions and the forces by which the produced pressure loads on active/passiive transducers. A fullly causal model is obtained by suitable regularization of the resonant frequencies associated the interior modes. Both models are compared to laser velocity measurements at the transducers’ centers of physical prototypes. This verifies the model up to frequencies where transducers break up into partial vibrations. Moreover, higher-order interior modes often become irrelevant, as at high frequencies the transducer mass starts to dominate.
	Reproduction of Multichannel Audio by Frame Loudspeaker Array Yuta Kakuzaki, Akio Ando [Abstract] Multichannel audio reproduces a sound field by loudspeakers surrounding a listener, which necessarily occupies the listening space. The purpose of this study is to reproduce the sound by loudspeakers placed around the image display, which is referred as ”frame loudspeaker array” in this paper. As the first step, we focused on the reproduction of the frontal channels of 22.2 multichannel by the array. Each channel signal of the multichannel audio was distributed into loudspeakers with the weighting coefficients that minimized the square error between the original sound field by the channel and the reproduced sound field by the array. The sound fields were represented by the tenth order spherical functions. The signal was passed through the auditory band pass filter and changed its amplitude and delay according to the weighting coefficient calculated for the center frequency of the filter. The band pass signals were then summed up to generate the full band input signal for loudspeakers. The informal listening test showed that this method could preserve the spatial feature of the original multichannel audio. The reproduction of channels other than the frontal channels of 22.2 multichannel will be studied in the next step.
	Three-Dimensional Spatial Active Noise Control Based on Kernel-Induced Sound Field Interpolation Hayato Ito, Shoichi Koyama, Natsuki Ueno, Hiroshi Saruwatari [Abstract] Active noise control (ANC) over a three-dimensional (3D) spatial region is proposed. Conventional multipoint ANC does not guarantee to reduce the noise between multiple discrete control points. Several attempts have been made to reduce the noise over the continuous target region. Most methods for spatial ANC found their basis on the spherical/cylindrical harmonic expansion of the sound field. Therefore, they can be applied only for arrays of simple geometries such as sphere and cylinder. We propose a 3D spatial ANC method based on sound field interpolation, which can be applied to arbitrary array geometries. The mathematical principle of the proposed method is based on our recent studies on kernel ridge regression for sound field interpolation. The difference in the resulting algorithm from the standard adaptive control appears only in the weighting matrix for an error signal, which can be computed in advance only by using the relative position of the microphones and the target region. Numerical simulation results indicated that a large regional noise reduction is achieved by the proposed method compared with conventional multipoint ANC.
	On in situ beamforming in an automotive cabin using a planar loudspeaker array Martin Bo Møller, Martin Olsen [Abstract] Compact loudspeaker arrays are often used in applications where control of the sound dispersion is required, e.g. in audio reproduction systems and aspects of sound field control. An example of the latter is sound zoning where the goal is to deliver individual audio content to one listener without disturbing another. Typical approaches are based on filter-and-sum beamforming applied to control the resulting directivity of the array. However, directivity is defined in free field while sound zones are generally of interest in listening spaces with reflective boundaries such as domestic rooms and automotive cabins. In order to address this discrepancy, in situ zone control can be utilized. This relies on measured transfer functions from each array element to the spatial control points defining the zones. Such methods increase the solution specificity and can lead to non-causal control filters. In the present paper, methods based on controlling the flow of acoustic energy are applied to a planar loudspeaker array located on the dashboard inside a car cabin. A comparison study including directivity and in situ zone control is presented, and the solutions are evaluated in terms of spatial robustness of the resulting sound zoning performance.
	Minimum Trap Separation for Acoustical Levitation Using Phased Ultrasonic Transducer Arrays Carl Andersson, Jens Ahrens [Abstract] Acoustic levitation for interactive visualizations is an emerging field which uses ultrasonic transducer arrays to induce radiation forces on small beads. To fully utilize the interactive and responsive aspect of the system each bead must be controllable freely in the interaction space without interfering with the other beads. When beads are placed far enough apart it is possible to design a sound field that traps all the beads at their desired positions, taking any potential limitations of the hardware into account. The underlying physics limit how close in space two taps can be without interfering with each other. In this paper, we investigate the minimum spacing required between two beads for them to successfully and independently levitate in acoustical traps. Multiple methods for the sound field design are considered and compared with regards to the overall separation required as well as the gracefulness in the breakdown region. Simulated results of the sound fields are compared and validated against measurements of real sound fields and real levitating beads.
	Spatial sound field reproduction using deep neural networks Thushara Abhayapala, Hanchi Chen [Abstract] Sparsity-based sound field reproduction algorithms often result in improved localization and larger reproduction region, but also lead to high computational cost. In this work, we present a novel approach for sparse reproduction, where a deep neural network (DNN) is trained to determine the optimal driving signals for a loudspeaker array, given the desired sound field coefficients as input. We show that when trained using the proposed method, the DNN-based algorithm can outperform existing Lasso-based algorithms in terms of noise sensitivity and computation speed.
	2.5D Local Wave Field Synthesis of a Virtual Plane Wave Using a Time Domain Representation of Spherical Harmonics Expansion Nara Hahn, Fieter Winter, Sascha Spors [Abstract] Wave field synthesis is a well established sound reproduction technique, where the wavefront of a desired sound field is physically recreated within an extended target area by using a large number of individually driven loudspeakers. For a non-continuous loudspeaker distribution, the synthesized sound field typically exhibits spatial aliasing artifacts which occur above an upper frequency limit. If the synthesis in a relatively small area is concerned, called local wave field synthesis, a substantial improvement of the physical accuracy can be achieved. In a recent local wave field synthesis method, proposed by the authors, the spatial bandwidth of the driving function is limited in order to reduce the spatial aliasing artifacts around a properly chosen expansion center. The present paper deals with the time domain realization of this method for a 2.5- dimensional setup. The driving function for a virtual plane wave is derived based on the time domain version of the spherical harmonics expansion, by exploiting the connection with the plane wave expansion. Due to the low computational cost, the proposed approach is well suited for real-time implementation. The spatial, temporal, and spectral properties of the synthesized sound fields are evaluated by numerical simulations.
	How the distance and radius of two circular loudspeaker arrays affect sound field reproductions and directivity controls Yi Ren, Yoichi Haneda [Abstract] We have proposed methods for sound field reproductions and directivity controls by using two parallel circular loudspeaker arrays (2CLA) so far. Results showed that 2CLAs have a higher potential than a circular loudspeaker array when reproducing a virtual sound source outside the array or producing a high directivity beam. Also, we found that the performance of the 2CLA varies with the distance between the two arrays and the radii of the two arrays. In this paper, we discuss how the two factors affect the performances of 2CLAs. We investigate the influences of either factor with the other fixed, via computer simulations. Results of sound field reproductions showed that the reproducing accuracy falls while the distance of the two array increases, while the radii of the 2CLA do not affect the results if no constraint has been set to filter gain. On the other hand, if filter gain is constrained to 0 dB, the reproducing gains while the radii increase. For directivity controls with filter gain constraints, the performance gets better when larger radii have been selected, while a large distance may make the beam wide.
	The EMPAC High-Resolution Modular Loudspeaker Array for Wave Field Synthesis Johannes Goebel [Abstract] In order to verify the perceptual potential and limitations of Wave Field Synthesis (WFS), a large system was built with loudspeakers placed as closely as possible. Design criteria focused on the frequencies dominant in the consonants found in human speech, which yield important localization cues, also in transients of instrumental sounds. The built system allows for undistorted wave field generation up to an aliasing frequency of 3.3 kHz, with experimental configurations allowing for observations of perceptual effects at aliasing frequencies between 700 Hz and around 6 kHz. Critical listening shows that raising the aliasing frequency to 6 kHz improves the spatial resolution substantially. A second generation with a loudspeaker distanced at less than 3 cm is under construction. Focused sound sources can be placed up to 10 m in front of the system. The current system’s 558 individually addressable loudspeakers are housed in 18 modules, each with 31 channels and an external subwoofer. The design allows for a high degree of flexibility in deployment, including vertical, overhead, and circular configurations, enabling the simulation of complex ”natural” acoustic environments and a diversity of applications in artistic, commercial and research-driven projects.
	A Field-Matching Method for Sound Field Synthesis for Large Scale Sound Reinforcement Systems Elena Shabalina [Abstract] There have been ongoing attempts to adapt existing Sound Field Synthesis methods for large scale sound reinforcement systems. In comparison to smaller indoor setups large scale systems have significant distances between secondary sources and often irregular loudspeaker layouts due to rigging conditions. These restrictions make the use of WFS-related unfeasible. In addition, prominent and frequency-dependent directivity patterns of concert loudspeakers obstruct the use of usual level and delay algorithms as they assume omnidirectional secondary sources. The paper proposes an alternative method based on a Field Matching technique that bypasses the above-mentioned restrictions. Despite the typically low aliasing frequency of a concert sound system layout the Field Matching approach was shown to provide an accurate localisation and a stable virtual source across the audience area including front seats. The paper presents a theoretical background and a Matlab simulation along with listening tests results. Limitation and challenges of the method for live sound applications are discussed.
	Adjoint-based sound reinforcement in the time domain Lewin Stein, Florian Straube, Jörn Sesterhenn, Stefan Weinzierl, Mathias Lemke [Abstract] Sound Reinforcement aims at synthesizing homogeneous sound fields for the entire audio bandwidth using Line Source Arrays. These arrays consist of electronically controlled loudspeaker cabinets with different tilt angles. The determination of the optimal geometric arrangement and electronic drive is an ill-posed inverse problem. Recently, the authors presented a novel adjoint-based approach which lives in the time domain as an alternative solution method to corresponding problems. We optimize acoustic sources to match a target sound field subject to an objective function. For that, we solve the Euler equations and the corresponding adjoint by means of computational aeroacoustic techniques. The method can determine appropriate driving functions and appropriate positions of the sources for the synthesis of the desired sound field. We will present validation examples including base flows, e.g. wind, and thermal stratification for a representative loudspeaker setup. It will be shown that the method provides reasonable driving functions also for these conditions.
	Full scale outdoor concert adaptive sound field control Jonas Brunskog, Franz Maria Heuchel, Diego Caviedes Nozal, Minho Song, Finn T. Agerkvist, Efren Fernandez-Grande, Enrico Gallo [Abstract] Outdoor musical events in urban environments is of large cultural importance, but is also a source of noise, especially for non-participating neighbors. In an attempt to solve this, an Adaptive Sound Field Control (ASFC) system is under development. The purpose is to mitigate low frequency noise annoyance and simultaneously enhance the musical experience of the audience. The system uses a secondary array of loudspeakers, canceling out the sound in a dark zone. To do this, accurate estimates of the transfer function between the loudspeakers and the dark zone is necessary. A static version of the system has been tested in three events: a controlled pre-test, the KappaFutur Festival in Turino, and Tivoli in Copenhagen. These deployments are reported and discussed, with special focus on the KappaFutur Festival, where noise measurements where conducted in the neighborhood, supplementing the insertion loss measurement the ASFC. The performance of the ASFC system is shown to strongly depend on the complexity of the setting. In the final stage, the system needs to be adaptive in order to adjust for changing climate conditions. The adaptive parts of the system is under development, and will be tested in a series of events in 2019.
	Adapting transfer functions to changes in atmospheric conditions for outdoor sound field control Franz Maria Heuchel, Diego Caviedes Nozal, Efren Fernandez-Grande, Jonas Brunskog, Finn T. Agerkvist, Peter Gerstoft [Abstract] The performance of active sound field control solutions is directly dependent on the accuracy of the measured acoustic transfer functions between the control loudspeakers and control areas. Outdoors, these transfer functions are affected by atmospheric conditions and their variation over time. In this work, we investigate strategies for transforming measured transfer functions to changing atmospheric conditions with the goal of adapting outdoor sound field control systems to such changes. Compared to active control methods based on adaptive filtering, such weather model based approaches do not rely on continuous sound pressure recordings inside the control area. We investigate different adaption strategies, one based on explicit delay compensation with effective speed of sound and an alternative based on machine learning. We train and test the adaption strategies against results of a large set of outdoor transfer function measurements in different atmospheric conditions.

Organization:

G. Elko, mh acoustics LLC

	Bayesian inference in direction of arrival analysis using spherical microphone arrays Ning Xiang, Stephen Weikel, Christopher Landschoot [Abstract] One challenging problem in acoustical applications using microphone arrays is determining the directions of arrival (DoAs) of multiple sound events. This work applies a unified Bayesian framework to address this problem in situations involving potentially multiple concurrent sound sources using spherical microphone arrays. It presents a two-level inferential problem of sound source enumeration and direction of arrival estimation. To this end, analytical models based on spherical harmonics are used to predict experimental data derived from spherical harmonics microphone arrays. The prediction is evaluated against the measured data in order to select the simplest such model that can adequately match the experimental data, thereby estimating first the number of sources, then their DoA information. This paper presents the analytic models, the two levels of Bayesian formulation, and analysis results to demonstrate the potential usefulness of this model-based Bayesian analysis for complex sound environments with potentially multiple concurrent sources. This paper discusses the impact of microphone array configurations upon DoA resolution. It additionally estimates performance of the method based on experimental results for simultaneous multiple sound sources, particularly in noisy or reverberant environments.
	Microphone array beamformer processing to reduce noise and reverberation Gary Elko, Jens Meyer, Eric Diethorn, Steven Backer, Tomas Gaensler [Abstract] Hands-free audio communication systems that are designed to allow audio and speech communication between remote parties are known to be sensitive to room reverberation and noise, especially when the source is distant from the microphone. One solution is to use an array of microphones to spatially filter the acoustic field so that the direct sound field from the talker is emphasized while reverberation and background noise is attenuated. It is well known that the maximum directional gain for an array of microphones is N^2 where N is the number of microphones. The maximum microphone array gain is only attainable with specific microphone geometries. Typical gains obtained with realizable microphone arrays are lower than this maximum and the loss can be significant. To increase the direct-to-reverberant ratio in a diffuse field by 20 dB a classical delay-sum beamformer (maximum directional gain of N) would need on the order of 100 microphones. Two algorithms are described that attempt to address the slow growth in directional gain as the number of microphones increase by replacing the standard linear microphone array beamformer with a parametric nonlinear multiplicative processing beamformer that uses multiple overlapping and non- overlapping beamformers to suppress noise and reverberation.
	The max-norm minimization in non-synchronous measurements Liang Yu, Yi Fan, Jerome Antoni, Haijun Wu, Quentin Leclere, Baihua Yuan, Wei Kang Jiang [Abstract] Acoustic imaging from non-synchronous measurements of microphone array can increase virtually the array aperture and microphone density by sequentially moving an arbitrary prototype array. It had attracted great interests recently, since it beyond the fundamental limitation of Nyquist-Shannon sampling theorem in the spatial domain (the working frequency that is determined by the size and microphone density of an array). The problem of archiving the non-synchronous measurements of microphone array boils down to a matrix completion of a block diagonal spectral matrix. In this paper, the max-norm is investigated for the low complexity modeling of the spectral matrix, and the corresponding Proximal-Point method for Max- norm based Spectral matrix completion (PPMSMC) is proposed for the matrix completion; Second, the rank, nuclear-norm and max-norm has been investigated as complexity measures in the context of spectral matrix completion, and their performances are compared at different setups of frequency range, number of sources and the increased virtual aperture size/density of microphone array. An experiment of a High-Speed Blender noise imaging is presented for validating the proposed methods.
	Microphone Array Method for Determining Noise Angular Energy Distribution on Building Envelopes Miodrag Stanojević, Miloš Bjelić, Dragana Šumarac Pavlović, Miomir Mijić, Tatjana Miljković [Abstract] A method is proposed for determining traffic noise angular energy distribution on building envelopes using the microphone array as a source localization tool. Measurement procedure is presented with the results obtained by a nonuniform planar array developed for this purpose with microphone positions optimized for traffic noise measurements. The results are obtained in two characteristic urban configurations - canyon and non- canyon streets. Differences are commented and their implications on the in-situ sound reduction index of the façade element are explained when the element is located in these distinct situations. The algorithms used for sound source localization are CB in the frequency domain, along with deconvolution algorithms DAMAS2 and CLEAN- SC, and their performance is compared. The paper addresses the problem of the computational complexity of the algorithms which results in long execution times and inhibits measurements on a larger scale. Various methods for algorithm optimization are explored by varying different parameters such as angular resolution, signal length, various algorithm specific parameters, etc. The sensitivity of output results on the variation of these parameters is analyzed. The goal is to create a time efficient measurement procedure which would enable the acquisition of a larger data set covering various urban terrain configurations.
	Modal Analysis for Damage Detection in Structures by Non-contact Measurements with a Commercial Microphone Array Olaf Bölke, Jan Heimann, Joquin Garcia [Abstract] When monitoring critical structures, fatigue fracture, deformations, holes and much more are cases of failure which must be detected at an early stage. Changes in the modal parameters (eigenfrequencies, damping ratios, and mode shapes) of the structure allow conclusions to be drawn about the extent and location of the deterioration, subject to appropriate preliminary examination. Conventional measurement methods (i.e. acceleration sensors, laser vibrometers, etc.) for vibration analysis have the disadvantage that they can either ”detune” the vibration modes due to their own weight and/or require a long measurement time due to their punctual measurement. In contrast, the use of a suitable microphone array allows the high-resolution acquisition of the entire surface vibration covered by the array. Thus, the modal parameters of interest are determined by measuring the pressure fluctuations in the near field of the structure. A commercial acoustic camera with 120 microphones (Fibonacci120 by gfai tech GmbH) is used for this purpose. On the basis of artificially generated failure cases (load fracture, inhomogeneities, etc.) on application-oriented, large-area structures, a method for the detection of failure cases using a microphone array is demonstrated. As a reference, simulated vibration analyses, as well as vibration analyses measured by laser vibrometers, are used.
	Analysis of the Sound Field in a Room Using Dictionary Learning Manuel Hahmann, Samuel Arturo Verburg, Efren Fernandez-Grande [Abstract] The sound field in a room is often modeled as a superposition of elementary waves, such as plane or spherical waves. These wave expansions provide a powerful means to interpolate or extrapolate the sound field within (and outside) the measurement domain. However, projecting the sound field of a large domain in a room into a planar or spherical wave base yields a high number of very elementary components. We examine the use of dictionary learning to find a set of more complex basis functions that are suitable to represent the sound field enclosed in a room. The resulting dictionary should be able to capture the dominant features of the sound field, and represent it using only a sparse set of functions. In this study, extensive measurements of the sound pressure in a room are obtained and used as a training set to learn a dictionary. We analyze the spatial properties of the learned dictionary, and compare it to simple elementary basis functions such as plane and spherical waves.
	Acoustic transfer admittance of cylindrical cavities in infrasonic frequency range Paul Vincent, Dominique Rodrigues, Franck Larsonnier, Cécile Guianvarc'H, Stéphane Durand [Abstract] Demand for calibration at infrasonic frequencies has emerged in response to earth monitoring problems. The primary standard for sound pressure is defined through the reciprocity calibration method specified in the International Electrotechnical Commission (IEC) Standard 61094-2:2009. This method is based on the use of closed couplers and is routinely applied by the National Metrology Institutes for a large frequency range; however, infrasonic frequencies below 2 Hz have not been explored until recently. The acoustic transfer admittance of the coupler, including the heat conduction effects of the fluid, must be modelled precisely to obtain accurate microphone sensitivity. IEC 61094-2:2009 provides two standardised solutions for the correction of heat conduction. However, researchers have noted significant deviations between these corrections at low frequencies in plane wave couplers, indicating that one or both techniques incorrectly calculate the influence of heat conduction. In this paper, the limitations of the standardised formulations at infrasonic frequencies are identified and two alternative solutions are proposed. An experiment is also reported, which highlights the discussed limitations of the standardised formulations for acoustic transfer admittance, while also demonstrating the validity of the proposed alternative formulations at frequencies down to 0.04 Hz.
	Alternating Least Squares-Based Joint Estimation of RETFs and PSDs for Multi-Channel Speech Enhancement Simon Doclo, Marvin Tammen, Ina Kodrasi [Abstract] The multi-channel Wiener Filter (MWF) is a well-known speech enhancement technique that can be used to improve speech quality and intelligibility of microphone signals recorded in noisy and reverberant environments. It is commonly assumed that (i) late reverberation and ambient noise can be modeled as a spatially diffuse sound field and (ii) the spatial coherence of the remaining noise is known a-priori. Using these assumptions, the MWF requires estimates of the relative early transfer functions (RETFs) of the target speaker and the power spectral densities (PSDs) of the target speaker, the diffuse and the remaining noise. Recently, we proposed a technique to jointly estimate these quantities by minimizing a model-based error matrix via an alternating least squares (ALS) method. In this contribution, we present extensive simulation results comparing the ALS method with a state-of-the-art reference method based on covariance whitening. We demonstrate the effectiveness of the ALS method in both stationary and dynamic acoustic scenarios by using the estimates in an MWF and evaluating its noise reduction and dereverberation performance with respect to the improvement in speech quality. Results show that the ALS method yields more accurate estimates than the reference method, especially in the presence of strong uncorrelated noise.

Organization:

K. Brandenburg, TU Ilmenau / Fraunhofer IDMT

Perceptual Aspects in Spatial Audio Processing Karlheinz Brandenburg, Bernhard Fiedler, Georg Fischer, Florian Klein, Annika Neidhardt, Christian Schneiderwind, Ulrike Sloma, Claudia Stirnat, Stephan Werner [Abstract] Spatial audio processing includes recording, modification and rendering of multichannel audio. In all these fields there is the choice of either a physical representation or of perceptual approaches trying to achieve a target perceived audio quality. Classical microphone techniques on one hand and wave field synthesis, higher order ambisonics or certain methods of binaural rendering for headphone reproduction on the other hand target a good physical representation of sound. As it is known today, especially in the case of sound reproduction a faithful physical recreation of the sound wave forms (”correct signal at the ear drums”) is neither necessary nor does it allow a fully authentic or even plausible reproduction of sound. 20 years ago, MPEG-4 standardized different modes for perception based versus physics based reproduction (called ”Perceptual approach to modify natural source” and ”Acoustic properties for physical based audio rendering”). In spatial rendering today, more and more the perceptual approach is used in state of the art systems. We give some examples of such rendering. The same distinction of physics based versus psychoacoustics (including cognitive effects) based rendering is used today for room simulation or artificial reverb systems. Perceptual aspects are at the heart of audio signal processing today.

Perceived Quality and Plausibility of Room Reverberation in VR Reproduction from Measured Images and Acoustics Luca Remaggi, Hansung Kim, Adrian Hilton, Philip Jackson [Abstract] Virtual reality (VR) systems have emerged as platforms for personal interactive audio-visual media experiences. In order to have a real world reference against which to evaluate the room acoustics reproduced within VR, methods are needed to compare the virtual experience with that of a real room. In this work, two pipelines were developed for evaluating room reverberation over headphones within VR, acoustically and visually. The acoustical pipeline involves parameterisation of B-format room impulse responses via the reverberant spatial audio object (RSAO) and subsequent binaural rendering. The optical pipeline uses stereo 360 images to reconstruct the room geometry and materials that are then translated into binaural via two popular VR software development kits. Formal audio- visual subjective tests were conducted to obtain ratings of perceived quality and plausibility, using hidden anchors, for six captured rooms. The results show a trade off between quality and interactivity, meanwhile the virtual acoustics were confirmed as plausible reproductions of the original rooms. Further work will examine methods to enhance the interactivity with realistic spatial room reverberation.

Organization:

F. Zotter, Inst. f. Elektr. Musik u. Akustik, KUG G. Tauböck, Österr. Akademie d. Wissenschaften, Inst. f. Schallforschung

On the Stability of Gabor Phase Retrieval Matthias Claus Wellershoff, Rima Alaifari [Abstract] Phase retrieval refers to the problem of recovering a signal from phaseless measurements. Gabor phase retrieval, in particular, is concerned with reconstruction from the absolute value of the Gabor transform (be it discrete or not) and has applications in the time-frequency analysis of audio signals. From a mathematical point of view, phase retrieval (from frame coefficients) is a challenging problem as it has been shown to be unstable in infinite dimensional Hilbert spaces and severely ill-conditioned in finite dimensional spaces. However, it has also been shown that one can relax the classical stability regime through a so-called semi-global phase reconstruction and obtain a stability result for phase retrieval from the continuous Gabor transform in this setting. Recently, we were able to adapt the semi-global regime to the discrete case and to prove a promising stability result from the discrete Gabor transform. In this contribution, we survey selected highlights from recent research on phase retrieval from frame coefficients with emphasis on phase retrieval from Gabor measurements. In particular, we review results on semi-global stability of phase retrieval in the infinite dimensional case and present our result on the finite dimensional setting.

Phase-Magnitude Relations and Phaseless Reconstruction for Time-Frequency and Time-Scale representations Nicki Holighaus, Zdenek Prusa, Günther Koliander [Abstract] We discuss the phase-magnitude (PM) relations for the short-time Fourier and wavelet transforms which are derived from analyticity of the respective transform, when an appropriate analysis window or wavelet is used. For the short-time Fourier transform, the PM relations were previously studied by Portnoff and later rediscovered by Auger et al. Recently, it was shown that these relations can be combined with an adaptive integration scheme to enable signal reconstruction from magnitude-only short-time Fourier measurements that often performs on par or better than more expensive, iterative algorithms for phaseless reconstruction. For wavelet transforms, similar relations hold that can be employed for phaseless reconstruction as well. We recall the phase-magnitude relations and the resulting algorithms in both settings and demonstrate their application in the audio domain.

Organization:

M. Cobos, Universitat de Valencia W. Kellermann, Friedrich-Alexander University Erlangen-Nürnberg

	Source Localization in Reverberant Rooms using Deep Learning and Microphone Arrays Hadrien Pujol, Éric Bavu, Alexandre Garcia [Abstract] Acoustic source localization is a well-studied topic in array signal processing, which could benefit from the emergence of data inference tools. We present our recent developments on the use of a Deep neural network fed with raw multichannel audio for sound source localization in reverberating environments. This paradigm allows to avoid the simplifying assumptions that most traditional localization methods incorporate. However, for an efficient training process, supervised machine learning algorithms rely on large sized, precisely labelled datasets. There is therefore a critical need to generate a large number of audio data recorded by microphone arrays in various environments.When the dataset is simulated either with numerical models or 3D sound-field synthesis, the physical validity is also critical. We therefore present an efficient tensor GPU-based computation of synthetic room impulse responses using fractional delays for image source models. We also present the use of physical 3D sound field synthesis for the learning process on microphone arrays. We discuss the advantages of this reproducible and semi-automated process, which allows to deal with arbitrary array geometries. We also analyze the localization performances of the proposed neural network fed with this dataset, which allows a precision as high as 1 degree in a reverberant room.
	Sound Quality Improvement for Speech Acquisition Based on Deep Learning and Harmonic Reconstruction with Laser Microphone Shoji Ueda, Kenta Iwai, Takahiro Fukumori, Takanobu Nishiura [Abstract] A laser microphone has recently been focused on acquiring the distant target speech without unnecessary sounds. It measures the vibration of the object near the target sound source by irradiating the object surface with the laser beam. However, the speech acquired with this microphone degrades the sound quality. For instance, the speech components at higher frequencies are attenuated by vibration characteristics of the object, and the speech is collapsed by stationary noise due to lower power of the laser beam from the object. To improve the sound quality of the degraded speech, deep neural network (DNN) has recently been proposed. It is trained by using a set of acoustic features extracted from the degraded speech and the clean speech. However, the speech components at higher frequencies are still attenuated after processed by DNN. Therefore, we propose the method to reconstruct the harmonic structure of the speech after processed by DNN. The notice point is that the harmonic structure composes the higher frequencies of the speech. The proposed method complements the spectral amplitudes on positive integer multiples of the fundamental frequency which are the components at harmonics. We evaluated the effectiveness of the proposed method though perceptual evaluation of speech quality.
	Performance Analysis of Audio Event Classification Using Deep Features under Adverse Acoustic Conditions Irene Martin-Morato, Maximo Cobos, Francesc Ferri, Javier Naranjo-Alcázar [Abstract] Audio event classification has been traditionally performed by extracting standard features based on human perception, such as Mel-frequency cepstral coefficients (MFCCs). However, the trend followed in the last years is primarily based on the information provided by deep features, which are extracted from the responses to complex input patterns learned within deep neural networks. These have been shown to obtain, in general, better performance than the hand-crafted ones. In fact, deep features are known to provide good generalization properties to classify events not seen during training, and can even be extracted from raw audio data. Since the captured audio data is highly dependent on the acoustic properties of the auditory scene, it is important to assess the impact that adverse acoustic conditions have in the final classification performance. In this paper, we analyze the robustness of deep features under controlled acoustic conditions by simulating different degrees of background noise, reverberation and segmentation errors, as well as in a real-life scenario where more than one audio event can be present at the same time. Results show an acute degradation in the performance given the background noise and segmentation errors, which suggest room for improvement in terms of robustness to different scenarios.
	Joint Separation, Dereverberation and Classification of multiple Sources Using Multichannel Variational Autoencoder with Auxiliary Classifier Shota Inoue, Hirokazu Kameoka, Li Li, Shoji Makino [Abstract] This paper proposes a unified approach for jointly solving separation, dereverberation, and classification of mixed sound sources in microphone array observations. The proposed method uses a frequency-wise convolutive mixture model to express the mixing process under highly reverberant environments and the auxiliary classifier conditional variational autoencoder (ACVAE) to model the complex spectrograms of underlying sources. The parameters of these models can be simultaneously estimated based on the same criterion by iteratively updating the separation matrix, the dereverberation filters, source-class probabilities and the power spectrogram of each source. Through experiments, we confirmed that the proposed method outperformed conventional methods in terms of the separation performance and the source classification rate.
	A Neural Network Approach to Broadband Beamforming Yugo Kuno, Bruno Masiero, Nilesh Madhu [Abstract] Beamforming techniques are commonly applied to signals captured by sensor arrays to enhance signals received from desired directions while reducing background noise and localised interference. Where the directions of the desired and interfering sources are known, this knowledge, combined with assumptions on the background noise characteristics, is used to derive the beamformer coefficients for each sensor. This derivation is usually obtained from an optimisation of the second-order statistics of the beamformer response, e.g., minimising the output energy while preserving the signals from desired directions. For broadband signals, like speech, the optimisation is often posed in the frequency domain. Beamformer coefficients are then independently derived for each discrete frequency, as an approximation of the true broadband response. Hereby, the complex inter- frequency interactions, e.g., due to windowing and spectral aliasing, are not modelled, leading to sub-optimal filter characteristics. These shortcomings can be overcome by data-driven approaches. By exploiting recent developments in deep learning, we propose a neural-network approach to generate a desired spatial response while implicitly accounting for the inter-frequency effects in the narrowband model. We test the proposed approach on synthetic and real-world data and compare the performance to standard (second-order) narrowband and broadband beamformers.

Organization:

B. Rafaely, Ben-Gurion University of the Negev G. Elko, mh acoustics LLC J. Meyer, mh acoustics LLC

	The uncertainty of room impulse response measurements Ingo Witew, Michael Vorländer [Abstract] The measurement of room impulse responses using two channel FFT correlation techniques is a standard approach to determine the acoustic properties of auditoria. Evaluating the uncertainty that is associated with this measurement method is challenging because the ”Guide to the Expression of Uncertainties in Measurements” originally requires establishing an algebraic relation between the system’s stationary in- and output quantities. 2-CH-FFT methods, however, are dynamic measurements as the impulse response is calculated from a microphone’s time domain signal, whose value at one time instant depends on the excitation’s time domain signal at previous time instants. Such measurement strategies don’t quite match the basic GUM approach. To reduce the complexity the dynamic in- and outputs of transfer function measurements are decomposed into quasi-stationary energy components. This allows investigating the uncertainty in line with the classical framework. On these grounds, general strategies are derived to determine the influence of background noise and system non-linearities in dynamic measurements. Second, these tools are used to consider the elements of the measurement chain and how they contribute to the uncertainty in measurements. The discussion concludes with an uncertainty budget for correlation technique measurements in rooms.
	Measurement of the four pole matrix of a sample in a transmission tube Rene Boonen [Abstract] A transmission tube has been developed to measure the four pole matrix of a sample of absorbing material. The electrical analogy of a sample in a transmission tube can be considered as a T-equivalent circuit of three impedances, two in series and one in parallel. As consequence, three measurements with three different closing impedances needs to be carried out. From these measurements, a set of equations will result from which the three impedances of the T-equivalent circuit can be determined in real and imaginary parts in terms of frequency. Once these impedances are known, any four pole matrix, such as the transfer or the scattering matrix, of the sample can be set up. The method has been validated by measurements of several sample combinations in the impedance tube. A good agreement has been established between transmission tube and impedance tube measurements in a range of 40Hz-4kHz.
	Ultrasonic Remote Sensing for Precision Agriculture Stuart Bradley, Mathew Legg [Abstract] Precision Agriculture makes use of quantitative measurements as input to sophistcated farm management software. For management of grazing animals, such as dairy cows, a key input is pasture biomass, so that pasture is not over-grazed nor have too much or too little application of fertilizer. Aside from destructively cutting, drying, and weighing the pasture, current measurement methods are slow and/or unreliable. We describe a new ultrasonic sensor which is compact and low- power, and which senses pasture properties remotely from a moving farm bike or UAV. The sensor comprises co-located log- spiral arrays of transmitters and receivers, giving high spatial resolution transverse to the propagation direction without excessive component counts. A pulsed, radar-like linear-FM chirp and matched filter gives high along-axis spatial resolution. With CLEAN image deconvolution, mm resolution is obtained through the pasture layer. This methodology allows detailed profiles of pasture density to be obtained at rates of 100 profiles per second. We present results retrieved via modelling of the complex acoustic scattering occurring in the pasture layer.
	PyTTa: Open Source Toolbox for Acoustic Measurements and Signal Processing William D'Andrea Fonseca, João Vitor Paes, Matheus Lazarin, Marcos Vinicius Reis, Paulo Mareze, Eric Brandão [Abstract] Python is a high-level programming language that has gained strength in the international community. This mainly occurs because it is open-source nature, that is, any user is entitled to its use, whether for scientific, commercial or hobby purposes. Another important feature is that Python is OS-independent. This means that no matter what OS the code was written, it can be run by any operating system. In addition, access to AD/DA acquisition interfaces allows interaction with real systems for their estimation and control. These features enable its use as a powerful signal processing tool, whether for acoustics or general purpose. The Federal University of Santa Maria (UFSM), in Brazil, has created a group of developers, who are working cooperatively on a toolbox for acoustic, audio and vibration signal processing. The project is named PyTTa, standing for Python in Technical Acoustics. It is object-oriented programmed, enabling the easy use of variables, classes, and functions. The collection of codes inside the toolbox creates an environment that facilitates the acquisition/measurement, pre/post- processing, I/O operations and plotting/documenting. This is a continuing work that relies on the collaborative effort to provide free access to information and work/research tools.
	A Theory for Stethoscope Acoustics Maximilian Nussbaumer, Leyre Troyas Martinez, Anurag Agarwal [Abstract] Maximising the signal to noise ratio while considering ergonomics and aesthetics is the key design challenge for modern stethoscopes. In order to optimise the design, there is a need for a well-validated model for the transfer function from a source within the chest to the output signal obtained. Most variants of the stethoscope are air-coupled sensors. Here we propose a new theory for the acoustics of this type of sensor, which takes into account the coupling between the sensor and the human chest. We have conducted a rigorous experimental campaign to characterise the transfer function of the chest-stethoscope system and have investigated the effects of key design parameters. Our data confirms traditional findings on the effects of bell geometry and diaphragm usage, but also highlights the importance of the coupling between the sensor and the chest, and reveals features of the transfer function that are not captured by existing models. Our model employs a transmission matrix formulation and discretises the system into lumped element components. It can be used to inform design choices for acoustic, electronic and dual-mode stethoscopes, opening up the possibility of an optimum design that maximises the signal to noise ratio for a desired application.
	Noise Intensity Prediction from Video Frames using Deep Convolutional Neural Networks Leonardo Oliveira Mazza, José Gabriel Rodríguez Carneiro Gomes, Julio Cesar Boscher Torres [Abstract] Some closed-circuit television (CCTV) systems do not have microphones. As a result, sound intensity information is not available in such systems. We present a method to generate traffic noise level estimates using solely video frames as input data. To that end, we trained a fully connected layer on top of VGG16 (pretrained with imagenet) using a dataset that was automatically generated by a single camera with a mono microphone pointing at a busy traffic crossroad with cars, trucks, and motorbikes. For neural network training from that dataset, color images are used as neural network inputs, and true average noise levels are used as neural network targets. The trained neural network successfully tracked trending noise levels with correlation 0.597 despite their blindness to the data temporal properties. These results suggest that average noise level targets are sufficient for convolutional neural networks to detect noise generating sources within a traffic scene.

Organization:

H. Führ, RWTH Aachen P. Balazs, Acoustics Research Institute, Austrian Academy of Sciences

Frame Theory for Psychoacoustics Peter Balazs, Nicki Holighaus, Thibaud Necciari, Diana Stoeva [Abstract] We will give a survey on the link between frame theory and signal processing tasks in psychoacoustics and biology. On the one side, the basic concepts of frame theory are presented. The goal is to reveal to scientists from psychoacoustics and biology how this mathematical theory can be a particularly useful conceptual background for them. Frame theory is a very active field in mathematics. It provides perfect reconstruction, stability, redundancy, and a signal-independent, linear inversion procedure, while still allowing a lot of flexibility. Those properties are in particular important to analyze any filter bank, thereby providing useful insight into its structure and properties. On the other side, basic psychoacoustic concepts are presented to stimulate mathematicians to apply their knowledge in this field. We will show motivating applications where the connection of theory with applications are particularly promising: perceptual irrelevance (i.e. how can we exclude those signal components that are not audible); as well as perceptually-motivated filter bank (i.e. how can we find the best time-frequency transform best adapted to human (or animal) hearing while still keep the frame properties mentioned above. This is a summary of the book chapter ’Frame Theory for Signal Processing in Psychoacoustics’.

A Localization Algorithm based on Head-Related Transfer Functions Maike Gerhard, Patrick Schillberg, Hermann Wagner, Hartmut Führ [Abstract] A localization algorithm based on head-related transfer functions (HRFTs) is introduced. The algorithm minimizes the distance to direction-dependent binaural signal spaces, in order to find an estimator of the actual sound source position. Numerical experiments with different target sounds (clicks, white noise, speech, rustling of leaves) and several signal-to-noise ratios, where the masker is unspatialized white noise, show that this projection algorithm outperforms template matching in nearly all experimental conditions. In particular, the algorithm is robust to changes of the emitted signal’s phase spectrum, unlike template matching. For white noise maskers, it is possible to compute SNR dependent estimates for the error probability in the task of discriminating two directions, based on the associated HRTFs. We present simulations that demonstrate the precision of the estimates. We show how these probabilities can be employed as a means to mathematically analyze HRTFs, in particular with the aim of predicting localization performance from the HRTF data set.

Organization:

G. Manley, Carl von Ossietzky University Oldenburg J. Christensen-Dalsgaard, University of Southern Denmark V. Singh, National Physical Laboratory, New Delhi

Evolution of Hearing in the Early tetrapods and Emergence of the Tympanic Middle Ear Jakob Christensen-Dalsgaard [Abstract] During the 120 my from the origin of tetrapods to the origin of tympanic ears in the Triassic, auditory sensitivity must have been based on non-tympanic, bone conduction mechanisms. However, bone conduction is a loose term for several different modes of stimulation of the inner ear. To understand hearing with a non-tympanic ear, I focus on the simplest: that sound translates the head region and that this vibration is transduced by the inner ear. The efficiency of translation of an object by sound is determined by ka (the product of wave number and radius) Comparison of the sensitivity of animals without middle ear (snakes, salamanders, earless frogs, lungfish) to sound and to vibrations of the skull shows that most sensitivity can be explained by translation. Interestingly, simple translation is also the mode of human low-frequency bone conduction sensitivity (for frequencies resulting in a ka>1). Translation by underwater sound is the mode of stimulation of most fishes. It is therefore a straightforward assumption that this was the mechanism of hearing also in the early tetrapods, and assessment of the sensitivity can inform us of the gradual changes leading to the tympanic ear, especially in relation to sensitivity and directionality.

Why did Solid Otoliths evolve in the Ears of Modern Bony Fishes? Tanja Schulz-Mirbach, Martin Plath, Friedrich Ladich, Martin Heß [Abstract] Modern bony fishes (Teleostei), which comprise almost one-half of all extant vertebrate species, show a remarkable diversity in ear morphology, including otolith shape. Fish bioacousticians are still puzzling over the question of why solid calcareous otoliths with species-specific shapes evolved, while most ”non-teleost” vertebrates possess numerous tiny otoconia. This question is linked to (i) when and how often a switch from loosely aggregated material towards a solid structure-and from apatite to calcium carbonate-occurred during the vertebrate radiation and (ii) whether those character shifts are related to altered hearing. We constructed a hypothetical framework for otolith evolution by compiling the available information on the structure of otoliths and otoconia in > 160 species covering all major vertebrate groups. Solid teleost otoliths may have initially evolved as a selectively neutral by-product of other key innovations, and the teleost-specific genome duplication event may have facilitated their subsequent diversification. Differences in otolith mass and shape might have enabled the perception of different ranges of acoustic information and may have evolved during the diversification of auditory abilities. In some teleost groups, otolith morphology may have co-evolved with ancillary hearing structures, especially if ears are closely connected to a swim/gas bladder.

Organization:

M. Wahlberg, University of Southern Denmark V. Singh, National Physical Laboratory, New Delhi

	Comparing pressure and intensity units for in-air and underwater hearing thresholds Magnus Wahlberg [Abstract] Hearing thresholds are usually reported in decibel units, which gives conveniently compressed numbers with biological implications. However, the decibel scale causes confusion when in-air thresholds are compared to underwater ones. Not only are the reference pressures of the in-air and underwater scales different, but the interpretation of the thresholds also depends on whether we assume the ear is detecting the integrated sound intensity or squared pressure (that is, if the acoustic impedance difference between the two media should be compensated for). Here, in-air and underwater hearing thresholds from the literature on toothed whales, seals, marine birds and turtles are compared, both using decibel scales and linear pressure and sound intensity units. The interpretation of how sensitive an animal is to sound in air and underwater critically depends on the choice of units used to report hearing thresholds. Therefore, great care must be taken choosing the adequate units in hearing studies.
	Evolution of Cetacean Underwater Hearing Mechanisms: Similarities and Differences between Archaeocetes, Mysticetes, and Odontocetes Sirpa Nummela [Abstract] Modern cetaceans have their origins in terrestrial mammals that started their aquatic life over 50 million years ago. The history of the earliest whales, the archaeocetes, is fairly well known, and in their skull anatomy one can follow their adaptation to life and hearing underwater, while the earliest ones of them were still able to live and hear on land, too. Mysticetes and odontocetes both stem from archaeocetes, but today these two groups live in very different acoustic worlds, having specialized in low and high frequencies, respectively. It is also apparent that they use different kinds of peripheral auditory mechanisms. It is currently under lively debate whether the ability to hear high frequencies was present already in late archaeocetes, with mysticetes then later specializing in low frequencies, or if the ability to hear high frequencies first appeared with the odontocetes. Sounds emitted by modern whales can be compared to their hearing characteristics, when these are available. Morphological data on the peripheral auditory structures can be used to predict hearing ranges of these animals. These questions will be reviewed and discussed in this presentation.
	The Tympanic Ear as an Efficient Underwater Sound Transducer Jakob Christensen-Dalsgaard [Abstract] All groups of tetrapods have members that adopt aquatic lifestyles with adaptations also of their auditory system. Water is a high pressure, low particle motion medium, and the consequence is that an efficient underwater ear is sensitive to sound pressure. It is often stated that underwater hearing can work efficiently without a middle ear apparatus by bone conduction, but the sensitivity of such an ear is limited by the very low particle motion in water. A comparison of tetrapods ranging from totally aquatic (the clawed frog Xenopus laevis) and mostly aquatic (the red-eared slider Trachemys scripta) to mostly terrestrial (the cormorant Phalacrocorax carbo sinensis) show similar features. All have tympanic middle ears with an air-filled middle ear cavity. The eardrum vibration peaks at the resonance frequency of the middle ear cavity air volume and the eardrum is modified (plate-like). In all three species, the lowest threshold to underwater sound is at this peak frequency and is around 80 dB re 1 µPa. Sensitivity to sound pressure is slightly lower in water than in air, making underwater hearing much more efficient in terms of sound energy. Consequently, the slightly modified tympanic ears of these species are efficient aquatic sound receivers.
	Late Evoked Potentials in Porpoises Speaks Against a Per- Click View of Echolocation, but Suggest Potential for Electrophysiological Audiograms at Low Frequencies Kristian Beedholm, Peter Teglberg Madsen, Michael Ladegaard, Peter Lloyd Tyack [Abstract] In echolocating animals, an open question is whether returning echoes are evaluated on a click-by-click basis or whether there is a running evaluation, incorporating several clicks in shaping the animal’s perception of the echo scene. During the buzz phase of small odontocetes, such as porpoises, inter click intervals (ICI) may be as low as 2ms, which would seem to preclude individual processing of clicks. We studied the low-frequency auditory evoked potentials (AEP) in a stationary porpoise presented with artificial clicks at ICIs of 0.5-1s. There is a persistent component in the AEPs occurring at latencies out to at least 150 ms, and it seems highly likely that this represents remnants of cortical processes, whereby 150ms should form a lower limit to the ICIs at which click-by-click processing of echoes could take place. Toothed whales do therefore not process echo information before emission of the next click. As an added bonus, we believe that the relatively slow components of the AEPs constituting the cortical contributions hold potential for assessing the audiogram of odontocetes electrophysiologically down to frequencies well below what is usually thought possible with traditional amplitude modulated tonal stimuli, often taken to be limited to frequencies above several kHz.
	The Cormorant Ear – Adapted to Underwater Hearing? Ole Næsbye Larsen, Magnus Wahlberg, Jakob Christensen-Dalsgaard [Abstract] Diving birds may spend several minutes underwater during foraging dives. However, surprisingly little is known about avian underwater hearing. We do not know their sensitivity or even if they respond to underwater sound. To help filling this gap we measured the audiograms of cormorants (Phalacrocorax carbo sinensis) and studied their ear anatomy. Wild-caught fledglings were anesthetized and their auditory brainstem response (ABR) to clicks and tone bursts was measured, first in an anechoic box in air and then in a large water-filled tank with their head and neck submerged 10 cm below the surface. The overall shape of their air-audiograms was like that reported for birds of the same size in air. The bandwidth and slopes of their audiograms were similar in air and water. However, in air the highest sensitivity was found at 2 kHz, whereas it was displaced towards lower frequencies underwater. These results suggest that cormorants have rather poor in-air hearing compared to similar-sized birds. Their underwater hearing sensitivity, however, is higher than what would have been expected for purely air-adapted ears. A possible reason for the poor in-air sensitivity is the special ear anatomy with the central eardrum shaped as a rigid piston like in turtles.
	Underwater Sound Localization using Internally Coupled Ears (ICE) Leo Van Hemmen, Anupam Prasad Vedurmudi [Abstract] Internally coupled ears (ICE), where an interaural cavity acoustically couples the eardrums, are an anatomical trait present in more than half of all terrestrial vertebrates. The superposition of outside and internal pressure on the two eardrums results in internal instead of interaural time and level differences, which are keys to sound localization. Although ICE is primarily a low-frequency terrestrial adaptation, the African clawed frog Xenopus laevis is a fully aquatic species with a distinct air-filled canal between the ears. In water, the speed of sound is four times that in air. Unlike terrestrial animals with ICE, the Xenopus interaural cavity is also medially connected to the lungs. By modeling the inflated lungs as a Helmholtz resonator, we demonstrate their effect in improving hearing in a low-frequency regime, while simultaneously enhancing sound localization in a disjoint high-frequency regime, corresponding to the frequency ranges of male advertisement calls. In conjunction with its unique plate-like eardrums, we show how Xenopus uses its ICE-like interaural coupling to generate considerable internal level differences between eardrum vibrations and thus overcomes the challenges of underwater sound-localization. Taken together, the two arguments of Helmholtz resonator and plate-like eardrums show the potency of ICE and are interpreted accordingly.

Organization:

H. Wagner J. Hildebrand J. Christensen-Dalsgaard, University of Southern Denmark

Effects of traffic noise, land use types and ecotones on the distribution of resident birds in a natural reserve in the Ruhr area, Germany Philipp Antoniou, Bryce Timothy Lawrence [Abstract] While being one of the most densely populated and formerly heavily industrialized regions of Germany, the Ruhr area still offers space for many diverse ecosystems whose preservation is regulated by law. Nevertheless auditory impairments, especially by traffic noise and due to spatial restrictions, are likely. This study examines the acoustic effects of traffic noise on the spatial distribution of birds in the natural reserve ëIm Siesack,’ Dortmund, guided by the research question: How does land use, road noise, and ecotones alter soundscapes and affect avian distribution?. To answer the research questions soundscape measurements were taken throughout the case study area (n=204) and analyzed utilizing five acoustic indices from the Soundecology package in R Studio. The results of the indices were used in a statistical analysis to investigate (1) correlations between the distance to roads and the amount of biophony and (2) the differences between biophonic soundscape indices in various land uses and ecotones. Significant correlations between distance to roads and the amount of biophony and significant differences between soundscapes in different land use types and ecotones were found. Consideration of the results as related to soundscapes and nature reserve policy in Germany are discussed.

Adaption to Bristish Standards to identify Construction Site Activity noise sources with the Potential to Cause Stress to Giant Pandas Murray Snaith, Alessandro Rodriguez, Brett Marmo [Abstract] Giant Pandas have a broader hearing range than humans and have higher sensitivity in different frequency bands. Giant Pandas are known to become stressed when subjected to undesirable noise, therefore noise which humans may deem unobtrusive has the potential to cause stress in Giant Pandas. As Giant Panda breeding programmes are taking place in zoos throughout the world it is important to understand the impact that anthropogenic noise and vibration has on these mammals and how it affects their wellbeing. This paper discusses the adaptation from international standards on noise and vibration to appropriately address the impact from construction site noise on Giant Pandas; as well as inform the acoustic considerations in the design of a new Giant Panda enclosure at Edinburgh Zoo in Scotland.

Organization:

J. Davy, RMIT University C. Guigou-Carter, Centre Scientifique et Technique du Bâtiment

	Comparison between prediction and measurement of sound attenuation associated to ventilation network elements Catherine Guigou-Carter, François Bessac, Simon Bailhache [Abstract] The AcouReVe Project (2015-2018) aimed to improve the knowledge and the quality of acoustic calculation in ventilation ductworks. Such calculations are based on simplified models and the main issue is the input data. For each component of the ductwork, acoustic insertion loss and/or sound generation due to air velocity has to be known. The present work concentrates on sound attenuation across different ventilation network elements, such as bends, branches and manifold. First a prediction method is described; it involves a multi-modal model pour elements associated to guided waves propagation and a finite elements model for complex elements. Measurements have been performed at CETIAT laboratory on different elements; the experimental setup is described along with the different evaluated configurations. The comparison between the measured results, the predicted results as well as the results from simple standard formulas applicable for the considered elements is presented and discussed in details.
	Sound Transmission through Double-glazed Window: Numerical and Experimental Analyses Chaima Soussi, Mathieu Aucejo, Walid Larbi, Jean-François Deü [Abstract] The domestic windows in the building facade has a significant role in sound insulation against outdoor airborne noise. According to the standards, the prediction of their acoustic performances is carried out in laboratory using two reverberant rooms. It’s well known that below the Schroeder frequency, the presence of these rooms can affect the measurements due to their modal behavior. In this work, various 3D vibroacoustic numerical configurations are studied to discuss the methods proposed by acoustic standards. In particular, we analyze the influence of (i) the presence of the acoustic cavities and (ii) the diffuseness of the acoustic field, on the sound reduction index of building elements. Moreover, due to the complexity of the problem (geometry, materials and boundary conditions), an accurate numerical model of the window should be developed. This model, based on the Finite Element Method, is built through a confrontation with experimental tests. For this, a modal analysis is carried out to characterize the vibrational behavior of the structure (eigenfrequencies, modal shapes and modal damping). Those parameters will be used to design an equivalent simplified model of the window, which will enable to carry out parametric analyses.
	Sound Transmission through Aluminum Framings of Window, Door and Façade Systems Yihe Huang, Fangliang Chen, Tejav Deganyar, Anselm Boehm, Rebeccah Grotemeyer [Abstract] Recent environmental regulations bring up higher standards on acoustic comfort. Mitigating the noise transmission from façade exterior to interior is rising as an important issue in building envelop designs. Modern façade systems consist of aluminum frames and high- performance multiple layer of insulating or laminated glass. Thus, the overall sound insulation of such systems depends not merely on glasses but might more on frames and the coupling between them. Particularly when the STC of glass reaches 40 or higher, sound path through the framing elements becomes more critical. However, this crucial issue has not received enough investigation and no satisfactory approach has been developed in the literature yet. To address this issue, the coupling between aluminum frames and glazing infill was carefully studied, and an extended model has been accordingly developed. A series of laboratory tests with different frames were conducted to validate the predictions. As an application of this research, a design-orientated simulating platform has been developed to provide quick estimations of the sound insulation performance of window and façade systems.
	Investigation of rebuild repeatability issues in laboratory impact testing John Loverde, Wayland Dong, Erik Holmgreen, Scott Bergquist [Abstract] Materials such as floor coverings, screeds, and resilient matting are commonly applied on top of a floor-ceiling assembly to improve the impact noise insulation. Since the test measures the insulation of the entire assembly, it can be difficult to isolate the effect of these top-side components from the base structural assembly. The authors previously reported on a laboratory test program to evaluate the repeatability in the laboratory when rebuilding assemblies [”Investigation of Extended Repeatability Conditions in Laboratories” ICSV25 (2018)]. As a continuation of that program, assemblies and products from historical tests were retested, and the resulting repeatability was poor. Investigation determined that the cause of the discrepancies was variation in the base assemblies, not the top-side materials. The techniques and results are discussed, and the effect on the interpretation of laboratory testing is evaluated.
	Uncertainty quantification of the diffuse sound field assumption in structure-borne sound radiation predictions Edwin Reynders, Pengchao Wang, Geert Lombaert, Cédric Van Hoorickx [Abstract] When predicting the radiation of structure-borne sound into a room, it is often assumed that the generated sound field is diffuse. A diffuse field is by definition a random field: it represents a conceptual ensemble of rooms with the same volume and total absorption, but otherwise any possible arrangement of boundaries and small objects that scatter incoming sound waves. Adopting a diffuse sound field model therefore inherently introduces uncertainty on the computed results. This uncertainty can be important, especially at the lower frequency end of the spectrum. In this work, practical formulas are derived for predicting not only the mean, but also the variance of energetic level quantities, such as the band-integrated spatially averaged sound pressure level, in a diffuse sound field caused by a mechanically excited structure. The formulas are first validated in a simulation study, and then applied for predicting the sound pressure level in a room caused by impact excitation of plate structures. Both the average sound pressure level and its standard deviation can be well predicted. This information is essential e.g. for judging at which frequencies the mean diffuse field estimate is sufficiently accurate for the application at hand.
	Numerical Investigation of the Reverberation Method for Measuring the Total Loss Factor of Plate-Like Structures Yasutomo Yamasaki, Naohisa Inoue, Tetsuya Sakuma [Abstract] The total loss factor measurement by the reverberation method is often performed to estimate the boundary conditions of vibration fields of plate-like structures. In order to clarify how measured values involve discrepancies from the true values, this paper presents a numerical model that simulates the reverberation method to determine the loss factor in a bending vibration system of a glass plate supported with an elastic material. As a reference, the total loss factor of a finite plate system is theoretically calculated based on the diffuse vibration field assumption, where the random-incidence vibration absorption coefficient on the support edge is given for the semi- infinite plate terminated by the mechanical and moment impedances. Subsequently, the finite element analysis is performed for a rectangular plate with the impedance boundary, and impulse responses are calculated with different arrangements of excitation and receiving points. Comparing the theoretical and numerical results, the validity of the measurement procedures is generally confirmed, however, it is demonstrated that the measured loss factors tend to become slightly higher than theoretical ones. Additionally, measured results on real window systems are discussed briefly.
	Simplified prediction of the vibration reduction indices of double wall junctions Arne Dijckmans, Lieven De Geetere, Charlotte Crispin [Abstract] Heavyweight cavity walls are commonly used as party wall between row houses and between dwellings in apartment buildings in Belgium. The double wall junctions complicate the prediction of flanking transmission. In this paper, empirical rules for the vibration reduction indices of rigid double wall T- and X-junctions are proposed. They are determined from Monte Carlo simulations carried out with a statistical model that was previously validated with laboratory measurements. In general, a double wall junction can be replaced by an equivalent single wall junction in flanking transmission models. The EN ISO 12354 approximation of a double wall T-junction by an equivalent single wall X-junction proves justified. For double wall X-junctions, the thickness of the equivalent single walls should be either the thickness of a single leaf or the total thickness of both double wall leafs, depending on the transmission path of interest.
	Sound Insulation Prediction of Single and Double CLT Panels Fredrik Ljunggren [Abstract] Cross laminated timber (CLT) as building material has rapidly gained in popularity in recent years, and thereby also the demand of engineering tools to predict the sound insulation through CLT walls. The sound reduction through walls of single and double panels is predicted and the results are compared with those found in the literature as well as with a series of measurements conducted in a test mock-up of two rooms. Various configurations are considered where the plate thickness, addition of plaster boards and cavity distance (in case of a double panel) are altered. The prediction of single panel walls is based upon an established model (by Sharp) for homogenous materials, based upon theory and empirical data. For optimized application to CLT constructions specifically, the model is modified to some extent. The double panel walls are analytically modelled under the presumptions of no mechanical connection between the two panels and a cavity filled with mineral wool. The model make use of the results from the developed model for individual panels. The accuracy of the predicted sound insulation is estimated to be within a couple of dB’s concerning the weighted sound reduction indexes from 100 and 50 Hz.
	Determining the airborne sound insulation improvement of thermal cladding systems in combination with heavyweight exterior walls Claire Churchill, Maximilian Neusser, Simon Hinterseer [Abstract] Prediction of the sound reduction index (SRI) of modern thermal insulation systems for exterior walls is considered. These products all have in common a lightweight insulating layer of primarily thermal function (which may consist of a porous or non-porous material), and a weatherproof finish; the systems measured in the laboratory included a grout and skim finish or a prefabricated cladding system. In all cases, a typical heavyweight exterior wall was cladded. Modern examples of such systems are WDVS (Wärmedämmverbundsystem) and curtain walls (Vorhangfassaden). In their simplest form, such cladding systems can be modelled as a spring-mounted mass. However, Austrian standards demand that the spring-mounted layer be combined with structural fixings. In the examined cases sparse distributions (up to 12 connections/m^2) of point connections were used. The SRI of simplified and typical thermal cladding systems (i.e. with and without structural fixings) were measured in an accredited laboratory facility and these results were compared with a calculation method to determine SRI. The possibility to use established methods to determine the coupling loss factors in such combined systems was considered. The structural wall is of heavyweight construction, therefore the SEA problem can be simplified, in many cases, to a primary path analysis.
	Determination and optimization of sound insulation capabilities of geometrically complex walls Elias Perras, Chuanzeng Zhang, Jihao Chen, Zhijiang Ji [Abstract] Several methods have been developed in the past to calculate the sound transmission in building acoustics. However, efficient determination and optimization of the sound insulation capabilities of geometrically complex walls like brick walls or layered wall structures remain a challenging task. In this work, a fully coupled fluid-structure interaction model, based on the frequency-domain spectral element method (FDSEM), is proposed to determine the sound insulation of arbitrary wall structures. A virtual sound measurement laboratory following the guidelines of ISO10140 is designed and used. The calculated frequency responses agree well with the experimental results. Therefore, this method can be used to evaluate the performance of the wall materials and structures in a cost-efficient way. Furthermore, contrary to other methods based on the plate theory, this method is more flexible and avoids many essential simplifications, which could restrict the applicability of the methods in the wall structure optimization. The present method is implemented in a multi-objective optimization procedure based on a genetic algorithm to optimize various wall structures (e.g. novel brick walls, multi-functional laminates), and the results are compared with that by other numerical and analytical methods.
	Prediction methods and evaluation of high sound insulations Maarten Luykx [Abstract] When consulting pop venues or cinema’s with critical spaces like dwellings nearby, high sound insulations will have to be realized. In an early stage of the design the related structural consequences will have to be known very stringently and reliable, because adaptations afterwards are usually not possible. In a recent situation with dwellings on top of a pop venue sound reductions of D’nT,A,house>95 dB had to be realized. Within the Peutz office several prediction methods for sound insulation are being used during the design, such as ISO 12354-2 and several own practical methods developed within Peutz, based on more than 65 years of experience. For this project specific high pressure vibration insulated mountings had to be designed and implemented. In order to evaluate the prediction methods used and to prove that the high sound insulations that were predicted have actually been realized, measurements after delivery have to be performed. In this case a special measurement setup was designed using 18 subwoofers and a noise level of 125 dB(A) house spectrum inside the pop venue, in order to measure and prove that extremely high sound insulations of more than D’nT,A,house=99 dB towards the dwellings had been realized.
	Comparison of transmission loss prediction using condensed equivalent plate models Arasan Uthayasuriyan, Fabien Chevillotte, Luc Jaouen, Dimitrios Chronopoulos, Emmanuel Gourdon [Abstract] Equivalent plate models are commonly used in order to simulate multilayer system as one single layer. This kind of approach enables for instance to reduce the mesh size and thus the number of degrees of freedom in finite element models. Usual condensed model rely on thin plate theory and are obviously limited when considering thick multilayers. In this work we present a comparison of the transmission loss prediction using different analytical models of solid layers: thin plate, thick plate (first order shear deformation) and 3D modelling (3D linear elasticity) . Doing so, analytical limits for the validity of thin and thick plate theories are derived from wavenumber analyses. Applications to real world cases are presented.

Organization:

J. Davy, RMIT University C. Guigou-Carter, Centre Scientifique et Technique du Bâtiment

	Components of Variation in Reverberation Time Measurement – Part 2: Field Testing Rooms of Heavyweight Construction Bill Whitfield [Abstract] This paper uses analysis of variance (ANOVA) and a specific design of experiment (DOE) and construction choice to isolate the component of variance associated with the part (or room) being measured along with individual contributions of measurement uncertainty from the measurement system and the person making the measurement. It demonstrates how the gauge repeatability and reproducibility (GRR) technique can be used to identify the variability in reverberation time measurement over the frequency range 100-3150Hz as and individual part of the process associated with field sound insulation testing in the UK. The experiment uses the measurement of reverberation time associated with the ëinterrupted source method’ from EN ISO 354: 2003.
	Airborne Sound Transmission Modeling of Walls based on Random Point Process Theory Cédric Van Hoorickx, Edwin Reynders [Abstract] A new method is presented to compute the diffuse sound transmission through walls that are either modeled deterministically or are taken to carry a diffuse vibration field. The fact that the natural frequencies can be considered to form a sequence of random points on the frequency axis makes this problem amenable to analysis by random point process theory. Furthermore, the method makes use of the fact that for any diffuse subsystem the statistics of its undamped eigenvalues and mode shapes saturate to universal distributions. The mode shapes saturate to Gaussian random fields, with a spatial covariance function depending only on wavelength and distance. The statistics of the local spacings between the eigenvalues saturate to those of a Gaussian orthogonal ensemble (GOE) matrix. This stochastic modelling approach is computationally very efficient as illustrated for a calcium silicate wall and a double glazing.
	Prediction of the sound transmission loss of two-dimensional periodic structures with a hybrid framework Carolina Decraene, Edwin Reynders, Geert Lombaert [Abstract] When considering a room-wall-room system to simulate a transmission suite in an acoustic laboratory, the direct sound transmission in the mid-frequency range can be obtained using a hybrid framework. The wall is then modelled deterministically as finite size effects and modal behaviour are still important, while the rooms carry a diffuse field and are modelled as stochastic subsystems. The finite element method is usually employed to compute the modal behaviour of the wall. At higher frequencies the computational cost then increases significantly as a finer mesh is required. For this reason, an alternative approach was recently developed, which allows to replace the finite element model by an analytical model for finite-sized thick and layered walls. However, the application of this approach is limited to layered structures such as sandwich panels or double glazing. In the present work, the efficient modelling of more complicated building elements which exhibit spatial periodicity, will be considered by invoking periodic structure theory. As a first application, the sound transmission loss of a brick wall is examined and compared to the predictions of a thick wall model with equivalent material properties and to experimental data.
	Acoustic topology optimization of the material distribution on a simply supported plate Jan Van Den Wyngaert, Mattias Schevenels, Edwin Reynders [Abstract] Lightweight plates typically have a low transmission loss due to their low mass. Since coincidence typically plays a role at high frequency, the TL can be increased with roughly 6 dB by doubling the thickness of the plate. By distributing the additional material in a non-uniform manner, a higher increase in TL can be achieved in a limited frequency band due to modal behavior. A good material distribution strategy is needed to reduce the transmitted sound in a certain frequency range. Here, the optimal material distribution at a single frequency and in a frequency band is investigated for a small simply-supported plate. The TL is modelled within the hybrid Finite-Elements- Statistical-Energy-Analysis framework to achieve a low computational cost while maintaining a high prediction accuracy. Using mass distribution, the TL is increased by shifting eigenfrequencies of the plate and creating stop bands. This is achieved by locally adding mass that disturbs the mode shapes by altering the bending stiffness of the plate. For a 1x1m2 PMMA-plate, an increase up to 33 dB in the TL is achieved for the single frequency case and between 3 and 30 dB for a 50 Hz frequency band.
	Adaptation of the Equivalent-Fluid Model to the Additively Manufactured Acoustic Porous Materials Kamil C. Opiela, Tomasz G. Zielinski [Abstract] Recent investigations show that the normal incidence sound absorption in 3D-printed rigid porous materials is eminently underestimated by numerical calculations using standard models. In this paper a universal amendment to the existing mathematical description of thermal dispersion and fluid flow inside rigid foams is proposed which takes account of the impact of the additive manufacturing technology on the acoustic properties of produced samples. The porous material with a motionless skeleton is conceptually substituted by an equivalent fluid with effective properties evaluated from the Johnson-Champoux-Allard-Pride-Lafarge model. The required macroscopic transport parameters are computed from the microstructural solutions using the hybrid approach. A cross-functional examination of the quality (shape consistency, representative surface roughness, etc.) of various periodic specimens obtained from additive manufacturing processes is additionally performed in order to link it to the results of the multiscale acoustic modelling. Based on this study, some of the transport parameters are changed depending on certain quantities reflecting the actual quality of a fabricated material. The developed correction has a considerable influence on the predicted value of the sound absorption coefficient such that the original discrepancies between experimental and numerical curves are significantly diminished.
	Sound Insulation of Monolithic or Laminated Single- and Double-Glazing Panels Fangliang Chen, Yihe Huang, Tejav Deganyar [Abstract] Sound transmission characteristics of single or multiple panels as sound insulation components in buildings have been studied extensively, and numerous models have been developed over the past few decades. Though most of them have been validated individually with specific test results in their developing phases, evidences showed that considerable discrepancy exit from one model to another. By applying many existing models onto glazing panels such as windows, doors or building facades, the authors found that most of them unable to accurately predict their sound insulation performance. Apparent discrepancy up to 3-5 dB in terms of OITC/STC index were even found between the predictions by commercial software and those well recognized test results. Obviously, more accurate theoretical models are highly demanded in industry to guide the design of real practice. To address this urgent need, revisits of most existing theoretical models were made by the authors, and extended models are accordingly developed and presented in this study in order to provide more accurate predictions on the acoustic performance of monolithic or laminated single- and double-glazing panels. Test results collected from literature as well as conducted by the authors are further provided to validate the proposed models.

Organization:

M. Schneider, Hochschule für Technik Stuttgart F. Morandi, University of Bologna B. Zeitler, Hochschule für Technik Stuttgart

	Floor Impact Sound Insulation of the Six-story Wood-frame Model Building Atsuo Hiramitsu, Ryuta Tomita, Hirakawa Susumu, Masayoshi Sato [Abstract] In Japan, the Act on the Promotion of the use of Wood for Public Buildings etc. was enforced in 2010. As a result, the number of constructions using the timber structure became more popular for the buildings. However, the performance of the floor impact sound insulation in the timber construction buildings is often concerned as worse than the concrete constructions. In this paper, an experiment was carried out on a newly built wood-frame six stories building to investigate the performance of the floor impact sound insulation on five different floors; CLT panel floor, stressed skin panel floor, LVL, I-joist floor and chord truss floor. The result showed that the stressed skins panel floor has the highest performance. The paper also reports the floor vibration performances and sound insulation of separation wall in the timber construction building, and proposes an improvement plan for the better floor impact insulation performance in the timber construction buildings.
	Hybrid joist floor constructions. Evaluation of measurement results. Anders Homb [Abstract] Lightweight building systems in general suffer from limited sound insulation, especially at low frequencies. Still, theoretical models have severe limitations regarding prediction of the impact sound insulation and the design is to a high extent based upon experience and measurements. Recommendations in Norway is including the frequency range down to 50 Hz, which means to include the spectrum adaptation term CI,50-2500 and C50-5000. A possible solution to improve the properties at low frequencies is to add mass and stiffness. Lightweight building systems combining wood and cement-based materials, we refer as hybrid constructions. The paper will present analysis of relevant measurement results of hybrid joist floor constructions from laboratory and field objects. The paper will discuss possible improvements of the floor, but also take the flanking transmission into consideration. From the ”Silent Timber Build project”, some relevant laboratory measurement data exist. Due to the impact sound insulation focus, some apartments with hybrid joist floors have been built and data collected. This paper is a part of a project at SINTEF Building & Infrastructure aiming to develop robust solutions also involving HVAC components inside the partition structure. Further progress in this project will include new laboratory measurements and a fire test.
	Application of elastic interlayers at junctions in massive timber buildings Stefan Schoenwald, Niko Kumer, Sebastian Wiederin, Norbert Bleicher, Bernhard Furrer [Abstract] The application of resilient interlayers at junctions in massive timber buildings was investigated in a parametric study in the Laboratory for Acoustics at Empa. Flanking sound transmission across a junction of a floor with an exterior wall was investigated for different construction details with and without interlayers, with rigid and decoupled connectors as well as with and without additional load. The performance of the interlayers is discussed by comparing the apparent airborne and impact sound insulation between two rooms one on top of the other. The benefit of the elastic interlayers and connection variants is hereby gauged against other additional measures, for example wall linings, suspended ceilings, floating floors required to achieve equal sound insulation and to fulfil acoustic requirements.
	Numerical simulation of CLT floors and comparison with empirical predictive models Marco Caniato, Federica Bettarello, Andrea Gasparella [Abstract] Wooden constructions are becoming more and more present in the global building market. They are built very quickly, combining structural and thermal performance in a single layering. They are industrially produced and then assembled on site to minimize building yard problems and increase construction speed. In this scenario, acoustics is often not addressed or is delegated to additional layers that are often inserted without a studied ratio. Many studies have been carried out to predict the acoustic behaviour of walls and floors using numerical or empirical methods. Recently, by means of validation with in situ measurements, the acoustic behaviour of raw wooden partitions has been defined with empirical or semi-empirical laws. In this work, a new method for the determination of the floor noise of raw CLT floors comes through the use of numerical simulation with transfer matrices. Then, the results are compared to the available empirical forecasting models and real measurements.
	Development of a quick and non-invasive measurement method for the extraction of the dispersion relation in CLT plates for the evaluation of the elastic parameters Arved Thies, Federica Morandi, Luca Barbaresi, Massimo Garai, Jörn Hübelt, Niko Kumer [Abstract] This contribution discusses a quick and non-invasive method for the extraction of dispersion relations in CLT panels, with the aim of providing an estimate of the elastic parameters of the plates starting from the fitting of the experimental dispersion curve to the theoretical models. Measurements were performed at the Stora Enso CLT mill on panels that have different thickness and orthotropic ratio. An instrumented hammer was used to excite the plate and accelerometers were used to acquire the impulse responses. Thanks to the reciprocity principle, the measurement points were kept fixed and the hitting point was moved; this allowed to reduce the measurement time and to decrease the number of accelerometers that must be attached. The dispersion curves were evaluated using the Phase Shift method and the experimental results were fitted with the theoretical model for acoustically thick plates; finally, the elastic parameters were estimated. The results give slightly low estimates of the Young’s modulus and a fairly good estimate of the shear modulus, while Poisson ratio that optimize the fitting are ”effective” values and not realistic values, as confirmed by recent literature. The precision of the hits is also analysed, while a reproducibility study is left for further developments.

Organization:

B. Rasmussen, SBi, Aalborg University (AAU-CPH) J. Jeongho, Fire Insurers Laboratories of Korea

	Survey method for rubber ball impact sound Jeong Jeongho [Abstract] Rubber ball impact sound was standardized in ISO standard as for precision and engineering method in ISO 10140 series and ISO 16283-2. Also, single number quantities for rubber ball impact sound is standardizing. In order to improve low-frequency impact sound isolation performance, quality control in a construction site is one of a necessary point to get reliable isolation performance. A survey method for airborne and light-weight impact sound was regulated in ISO 10052. However, a survey method for rubber ball impact needs to be proposed. To proposed survey method for rubber ball impact sound, the proposed method should be correlated well with measurement results which were measured with precision or engineering method. In this study, field measurements were conducted using KS F 2810-2 with five impact positions and five receiving points in Korean apartment buildings and the results were compared with survey method results which are extracted from the field measurement results.
	Rain Noise Brian Donohue, John Pearse [Abstract] In this paper we discuss the requirements of the generic international standard (ISO, BS EN ISO 10140) for testing of sound transmission through sample roofs exposed to simulated rainfall and of lessons learned during a recent test program. The test data forms the basis for calculating in-situ sound levels in rooms beneath the roof and we discuss the differences in sound produced by simulated rain to that of natural rain. The differences in impact velocity and raindrop distribution between simulated and natural rain are key factors that are not addressed by the Standard. In addition, an optional normalisation test using a pane of glass is included, for the explicit comparison of products tested and as quality control for test laboratories. As this test returns a lower sound intensity value it has wrongly been used by manufacturers as the basis for calculating room sound levels. The Standard does not specify whether the normalisation test should be carried out as a skylight or as glazing but the two tests have different requirements. Being optional and intended for inter-lab comparison suggests that the normalisation data should not be released to clients as it is misleading and thus should be excluded from reporting.
	Ratings and classifications for high-frequency impact noise isolation Wayland Dong, John Loverde, Jochen Scheck [Abstract] A two-rating method of evaluating impact noise isolation has been proposed by two of the authors [LoVerde and Dong, J. Acoust. Soc. Am. 141, 428 (2017)], in which low and high-frequency components are evaluated independently. Numerous other researchers have investigated the issue, and although the details differ, there has been some convergence of ratings and classifications for low-frequency impact noise. For the high frequencies, most have assumed that the existing IIC/Ln,w ratings are appropriate, while the authors have proposed high-frequency ratings which are based on the existing ratings but with a limited frequency range of 400-3150 Hz. Analysis of impact data from floor assemblies common to North America shows that the proposed ratings evaluate high-frequency impact sources and mitigation measures better than the existing ratings. The analysis is extended to include data from German floor systems and heavyweight and lightweight stairs. The ratings are compared and possible classifications are suggested.
	Airborne sound insulation in dwellings – single numbers weighted from 50-3150 Hz correlated to Swedish questionaire surveys Christian Simmons, Fredrik Ljunggren [Abstract] In course of work with the draft standard for sound classification of dwelling (ISO/DIS 19488), it was vibrantly debated whether 50 or 100 Hz is the proper lower frequency limit when evaluating airborne sound insulation between dwellings. Although 100 Hz is used in regulations in most countries, there is an ongoing interest in including lower frequencies. In Sweden, evaluation from 50 Hz became mandatory already in 1999 wherefore unique long term experiences from the field are available. In this paper, field measurements are compared to subjective ratings given by the residents. The questionnaire surveys were distributed in total 46 building objects with light weight or heavy concrete walls and floors. Several single number quantities in ISO 717-1, were correlated against the resident’s ratings. The statistical evidence for a 50 Hz limit was found to be small considering all buildings, but analyzing lightweight buildings only show an importance of using 50 Hz. We recommend to measure from 50 Hz when goodsound protection against a broad variety of sound sources is at stake, including music at low frequencies. If protection against disturbing speech is the only concern, measuring from 100 Hz is enough.
	Is noise annoyance from neighbours in multi-storey housing associated with fatigue and sleeping problems? Birgit Rasmussen, Ola Ekholm [Abstract] Danish Health and Morbidity Surveys including questions about neighbour noise and traffic noise have been carried out each 3-5 years since year 2000, the latest one in 2017. In all five surveys, the neighbour noise annoyance was higher than annoyance from traffic noise. In 2017, 14,022 respondents (response rate 56%) completed the self-administered questionnaire, out of which 3,893 adult Danes lived in multi-storey houses. Noise annoyance was assessed by asking the respondent about noise annoyance from neighbours and traffic, respectively, in their home during the past two weeks. Fatigue and sleeping problems during the past two weeks were assessed with a checklist that included eight different health outcomes (including the two mentioned). In total, 36% reported having been very/slightly bothered by neighbour noise and 22% by traffic noise. Noise annoyance from neighbours was strongly associated with both health outcomes. For example, individuals being very annoyed by noise from neighbours had 2.91 times higher odds of being very bothered by fatigue. Similar associations were also observed for traffic noise annoyance. Although, causality cannot be established in the present cross-sectional study, it is concluded that noise annoyance from neighbours is strongly associated with both fatigue and sleeping problems.
	Perceived noises in your residence: which one annoys the most? Talita Pozzer, Rodrigo Silva Motta, Elaine Lemos, Débora Leite, Iara Cunha [Abstract] The Brazilian standard NBR 15575 establishes performance criteria for residential buildings, including acoustic performance requirements for airborne sound insulation of internal walls, facades and floors, as well as for impact sound level of floors. The standard also specifies non-mandatory sound pressure levels in the dwelling from service equipment for informative purposes. This study aims to investigate which type of noise source is the most uncomfortable for Brazilian dwellings occupants. In order to reach the proposed objective, a preliminary questionnaire for collecting sound source descriptors was carried out. This step aims to gather free-form descriptive words of the sound sources from subjects that are not familiarized with acoustic terms. After the annotation analysis, a questionnaire was applied using a ranking scale. The selected descriptive words were also correlated to the noise sources mentioned in the standard. It is expected that this work may give input to the ABNT NBR 15575 systematic revision, that kicked off in 2018.
	Subjective Evaluation of Acoustic Quality Classes in Dwelling Seda Kulak, Nurgun Bayazit [Abstract] The aim of the study was to evaluate the acoustic quality classes subjectively and compare the results with objective metrics. Interior noise level in a dwelling is described in the Regulation on Protection of Buildings Against Noise in Turkey with five classes, three of them (B, C, D classes) was simulated via laboratory listening test experiment. 30 normal hearing participants rated the most annoying neighbor noise is determined with previous study and recorded in selected household. The subjects also rated three classes in terms of loudness and annoyance characteristics of the sounds transmitted through 3 different walls (masonry wall, aerated concrete wall, dry wall) or 2 different floors (reinforced concrete slab, hollow tile floor slab) types which were simulated by audio filters. Computer based interface was developed to perform the experiment. The sound level in a headphone was calibrated by using head&torso simulator. The results were statistically analyzed and subjective evaluation was correlated with physical measurement of sound metrics.
	Comparison of Acoustical Performance and Subjective Evaluation in Residential Buildings Ayca Sentop, Nurgun Bayazit [Abstract] The ”Regulation on Protection of Buildings against Noise” was issued by the Turkish Ministry of Environment and Urbanization in May 2017. The regulation was put into force partially in May 2018 obliging all new buildings to follow acoustical requirements. In May 2019, acoustical project by an acoustical expert will be obligatory for large scale or multifunctional buildings. The regulation aims to introduce a scheme for acoustical performance evaluation, ranging from A to F following the EU/COST TU 0901 project outcomes and ISO-CD19488. This study aims to demonstrate the application of the regulatory requirements with field measurements and a survey. Airborne and impact sound insulation, indoor noise level, mechanical noise level and reverberation time were measured in 6 residential buildings and their performance were analysed according to the classification scheme. A survey was conducted with the residents of the buildings. The results were compared and the applicability of the regulation was discussed.
	Sound insulation between dwellings – Comparison of national requirements in Europe and interaction with acoustic classification schemes Birgit Rasmussen [Abstract] Most European countries specify limit values for airborne and impact sound insulation in housing. The requirements are expressed by descriptors defined in ISO standards implemented also as European and national standards. A comparative study investigating the regulatory requirements for airborne and impact sound insulation between dwellings in 35 countries in Europe has been carried out. The comparison shows considerable differences in terms of descriptors, frequency range and level of requirements. Sufficient sound insulation between dwellings is important to protect against noise from neighbours and to provide privacy and possibilities for activities without causing annoyance. In addition to acoustic regulations, about one-third of the countries have developed acoustic classification schemes with classes reflecting different levels of acoustical comfort/protection. In some countries regulations point to a specific acoustic class, but are in other countries independent. The interaction will be described. All schemes have minimum two classes above regulations, thus enabling specification of higher acoustic quality than regulations, but typically only one class below regulations, implying that major parts of older housing cannot be classified. It is concluded that acoustic classification schemes should preferably include classes covering all existing housing, independent from construction year, thus allowing acoustic labeling in analogy to energy labeling.
	Acoustic regulation in hospitals – Interior acoustics improving the recovery of patients Georg Schöner [Abstract] Busy environments are bound to be noisy and hospitals are no exception. The rampant noise found in modern hospitals are caused by a variety of sources such as patients, staff, and medical alerts, making hospitals unable to meet the desirable acoustic levels. Evidence-Based Design has the ability to reduce treatment times, the consumption of medication by patients, as well as create an environment that supports better sleeping patterns and calmness. For example, a study of 416 patients has shown that quieter environments can reduce hospitalization periods. Major European countries, such as NL, IT, UK, still miss any acoustic regulation for the healthcare sector. Other countries, such as FR, GER, ES, PL, FI and the Scandinavian countries have specific regulation for hospitals in place. This contribution will analyze the details for the requirements in 8 European countries. In addition the discussion will get initiated on what a meaningful ”common denominator” could look like for countries with neither interior acoustic guidance nor regulation in place.

Organization:

B. Rasmussen, SBi, Aalborg University (AAU-CPH) J. Jeongho, Fire Insurers Laboratories of Korea

	The New Italian standard UNI 11532 on acoustics for schools Arianna Astolfi, Linda Parati, Dario D'Orazio, Massimo Garai [Abstract] In Italy, the new UNI 11532 standard introduces requirements for acoustic comfort in different building typologies such as schools, offices and hospitals. In March 2018 the UNI 11532 - Part 1 ”General requirements” was published, which includes the main descriptors related to noise, room acoustics and speech intelligibility to be considered for the design and verification phases. The drafting of Part 2, dedicated to schools, is now nearing completion. This standard is voluntary for private constructions, but mandatory in the public sphere following the issue of the Ministerial Decree ”Minimum Environmental Criteria for the awarding of design and construction services for new construction, renovation and maintenance of public buildings” (Criteri Ambientali Minimi or ”CAM”), dated 11 October 2017. The Decree explicitly refers to the UNI 11532 standard for the requirements to be guaranteed in public procurement. The underlying principles and criteria of the new standard, as well as examples of application and comparisons with other standards in Europe will be presented and discussed in this work. Good acoustics is one of the main requirements from kindergarten to university, where information is transmitted from teachers to students mainly orally, and where learning is strictly related to the quality of this communication.
	New Finnish building acoustic regulation Mikko Kylliäinen, Ari Saarinen [Abstract] New building acoustic regulation and related guidelines took effect in Finland recently. The statutes include provisions concerning sound insulation, noise and vibration abatement and acoustic conditions necessary for the construction of a new building, repair and alteration of a building and the changing of the intended use of a building. The essential technical characteristics enacted concern health and wellbeing in the manner required by the intended use of the building. The purpose of the guidelines is to clarify and facilitate compliance with the requirements as the statute directs. The guidelines provide instructions on the procedures for design and verification. The statutes apply to buildings with dwellings or accommodation or patient rooms, facilities for teaching, meeting, eating, treatment, recreation or physical exercise, or office spaces.
	Acoustic requirements vs experienced sound in wood structures Klas Hagberg [Abstract] Wood, in particular CLT, is increasing as structural material in buildings, improving the conditions for circular economy within the construction sector. However, structural wood is still creating doubts amongst many developers and one issue is the difficulty to easily fulfil equal sound insulation requirements as we are used to from the history with concrete. The difficulty increases when the buildings are high. To secure the Swedish sound insulation class B, either the walls have to be very thick or elastomers have to be installed between the different storeys. Often high design margins are needed since the real efficiency of the layers is unknown when the wall elements are tightly connected due to static reasons. Therefore, it is the duty of acousticians to ensure that the requirements we are stating in the design process really correspond to the annoyance of the habitants for different building categories. We should not lower the overall quality but instead refine the requirements to be more spot on in each single situation. Therefore, increased knowledge is necessary, in particular regarding annoyance in different housing categories. This paper gives examples of the challenges that we need to take into account for the future development of wood technology.
	The increment in cost and other parameters to upgrade quality classes in sound insulation Bilge San Ozbilen, Nurgun Bayazit [Abstract] Providing comfortable environments for users is the main objective for architects at the design process today. Physical performance parameters of building elements such as sound insulation, thermal insulation, resistance to fire and moisture are evaluated primarily for user comfort. However, in design phase elements’ costs, thicknesses and weights are considered as well. To ensure sound insulation, according to Acoustic Regulation of Turkey, minimum quality class of C should be provided. The practices like adding layer or increasing weight to upgrade quality class from C to A or B, generally are in confict with the cost criterion. In this study, 366 number of internal, 252 number of external masonry wall alternatives made of brick and autoclaved aerated concrete blocks are evaluated. The optimal wall types for each classes are selected by TOPSIS (Technique for Order Preference by Similarity to Ideal Solution) method. The weighing coefficients, representing importance, are given as sound insulation > cost > weight=thickness= thermal transmittance. This paper presents results of the calculation of required increments as percentages in cost, thickness and weight to upgrade quality classes.

Organization:

A. Mayr, Technical University of Applied Sciences Rosenheim S. Schoenwald, Empa Materials and Technology

	Single rig two-stage method for characterizing structure-borne sound sources in buildings Barry Marshall Gibbs [Abstract] Two source quantities are required to predict the structure- borne power into lightweight buildings: source mobility and free velocity or blocked force. The average source quantities can be measured using the reception plate method (RPM). A thin reception plate gives an estimate of the sum-square free velocity. A thick reception plate gives an estimate of the sum- square blocked force. The ratio of these quantities approximates the square of the average source point mobilities. In this proposal, only one reception plate is required. The transfer mobilities from several remote points to a single measurement centre are recorded. The source is attached about the measurement centre, and the transfer mobilities re-recorded. The ratio of the transfer mobilities gives an estimate of the source mobility. The source is set into operation and the velocities at the remote points recorded. For each remote point, the recorded velocity and previously estimated source mobility gives an estimate of the source free velocity. In the proposed method: there is no need to change the installation conditions; knowledge of source-receiver mobility ratio is not required; the reception plate can be relatively thin/light; the method automatically collapses the source data to equivalent single quantities.
	Experimental investigation of a single reception plate method to obtain two source quantities required to predict structure-borne sound transmission in buildings Fabian Schöpfer, Andreas Mayr, Ulrich Schanda [Abstract] To predict the structure-borne sound power into lightweight buildings and other structures, two source quantities are required: the mobility and the free velocity or the blocked force. Using the reception plate method (RPM), the average source quantities can be measured. The combination of measurements on a thick and thin reception plate, the so-called two-stage method, allows the determination of all required quantities using the RPM. The previous paper introduced a proposal that requires only one reception plate for this task. This paper shows an experimental case study that applies this approach. As reference, the free velocity and the source mobility are determined using measurements on the source approximating free conditions.
	Prediction of Maximum Fast Time-Weighted Sound Pressure Levels from Time-Varying Structure-Borne Sound Sources in Heavyweight Buildings Steffi Reinhold, Carl Hopkins, Gary Seiffert [Abstract] At present the prediction of the sound pressure response in a nearby room due to structure-borne sound excitation from building machinery is limited to sources running under steady-state conditions according to the method described in EN 12354-5. As machinery noise also needs to be assessed under running conditions which are not steady-state, some regulations in European countries on installation noise also set requirements in the receiving room based on maximum Fast time-weighted sound pressure levels. In this paper investigations were carried out on a laboratory heavyweight reception plate to quantify the structure-borne sound power of different idealised time-varying signals using equivalent continuous levels over short time periods for input into the prediction model. Empirical corrections to previous tests are then applied to estimate maximum Fast time-weighted sound pressure levels from predicted short equivalent continuous vibration levels in heavyweight buildings. The maximum Fast time-weighted sound pressure level prediction results are validated against in-situ measurements in the building-like situation with close agreement between predictions and measurements.
	In-plane excitation of reception plates according to DIN EN 15657:2017 Ulrich Schanda, Michael Hoßfeld, Fabian Schöpfer, Andreas Mayr [Abstract] DIN EN 15657:2017-10 describes methods for the determination of characteristic structure-borne sound power of building service equipment. The so-called indirect method is based on the use of reception plates. The method is robust for the case when the mobility of the source (building service equipment) is significantly higher than the mobility of the receiving plate, which was shown in earlier investigations . In those studies the building sources were all mounted on the plane of the reception plate. Therefore the excitation of the reception plate happens out of plane. In practice however there are situations which require mounting in wall openings, for example roller shutters in window openings or doors. In these cases the excitation happens in the soffit of the structures, the excitation takes place in-plane. The assumption then is that primarily longitudinal waves rather than bending waves are excited in the receiving building elements. This mounting situation was experimentally investigated on a specific reception plate in the lab. In this paper first results will be presented and discussed with the focus on the determination of the characteristic structure-borne sound power of those sources which could be used as input data for prediction models.
	Extraction of high contributing sound of Air Handling Unit and noise reduction using transfer path analysis Takuya Hirao, Yosuke Miyamoto, Junji Yoshida [Abstract] Air Handling Unit (AHU) is generally built in a machine room of a building and controls the temperature in multiple office spaces. The motors of AHU generate sound and vibration and have a possibility to cause noise subject in office. Accordingly, it is necessary to extract high contributing part of the AHU and apply intensive countermeasures to the part quickly. In this study, we attempted to reduce the noise by applying the operational TPA to the simple AHU model. In the test, the radiated sound in front of the model was measured and the frequency band which increases the over all noise was extracted. In addition, operational TPA was applied to find out the main contributor to the radiated noise. As the result, the noise at 200 Hz band was large and the motor radiated noise was found to be main contributor. Furthermore, the noise was increased in the unit case by the cavity resonance. Then, the inside volume of the AHU was separated using a plastic plate for the reduction of the motor noise in the unit. The result shows that the unit radiated noise at the frequency band could be decreased well.
	Experimental Validation of a Finite Element Model for a Heavy Impact from the Standard Rubber Ball on a Timber Floor Xiaoxue Shen, Carl Hopkins [Abstract] In lightweight buildings there can be problems due to heavy impacts, such as from footsteps in bare feet or children jumping. For this reason the standard rubber ball is used in measurement standards to assess heavy impacts. In this paper, experimental work on a mock-up timber floor is used to provide benchmark data to validate a model using Finite Element Methods (FEM) of the standard rubber ball exciting a timber floor. Experimental Modal Analysis was initially used to extract damping information and the Modal Assurance Criterion was used to confirm that the FEM model of the timber floor was reasonable. A FEM model was then used to predict the time-domain response of the timber floor when excited by a single impact from the standard rubber ball which was included in the FEM model. Comparison of these FEM predictions with experimental data was assessed using the Frequency Domain Assurance Criterion which indicated reasonable agreement with the experimental results.
	Structure-borne sound sources in timber buildings – prediction of machinery noise using measured transmission functions Fabian Schöpfer, Andreas Mayr, Ulrich Schanda [Abstract] Prediction of machinery noise in buildings involves two stages: (a) the emission of structure-borne sound and (b) the transmission (i.e. the propagation) across the building structure. This paper concerns the latter in timber buildings. In previous work a practical approach to predict the equipment sound pressure level using measured transmission functions was presented. A research project conducted at the Technical University of Applied Sciences Rosenheim pursued this approach by measuring a range of transmission functions in various timber buildings. In total a data set of 120 transmission functions measured in 19 timber buildings is now available. This paper gives a survey of this data by grouping similar constructions and transmission paths. The approach can be used to derive average transmission functions for similar situations. Those could potentially be provided in a catalogue giving data in one-third-octave bands. The application of this empirical prediction method is presented in case studies using fictive and real structure-borne sound sources in timber buildings.

Organization:

A. Santoni, Engineering Department, University of Ferrara C. Hoeller, OTH Regensburg University of Applied Sciences P. Fausti, Engineering Department, University of Ferrara

	Measurement of the bending stiffness of masonry walls by using time-frequency analysis: Practical considerations and validation Charlotte Crispin, Christian Mertens [Abstract] The bending stiffness is one of the parameters that determines the mechanical behaviour of a plate and therefore influences the acoustic performance of building elements. Besides, it appears in the prediction models for calculating the sound reduction index, R, or recently the vibration reduction index, Kij. Its accurate determination is thus essential in the field of acoustics. The Atomic Energy Research Institute of Korea has proposed an interesting measurement technique to determine this parameter in-situ. This technique, which uses the time-frequency analysis, has been successfully applied to thin plates. However, this paper shows that corrections should be used for thick walls. Some practical considerations are given and some measurements results for masonry walls are presented.
	Benchmarking of Methods for the Identification of Flexural Wavenumbers in Wooden Plates Federica Morandi, Olivier Robin, Luca Barbaresi, Massimo Garai, Noureddine Atalla, Nicolas Quaegebeur, Patrice Masson [Abstract] This paper investigates and compares three methods for the identification of the flexural wavenumbers in thin laminated wood panels with the aim of verifying under which conditions each setup can provide accurate results. Laboratory measurements are performed on a plywood panel, an inhomogeneous and orthotropic plate, and the propagation characteristics of flexural waves are characterized along five radial directions. The methods considered for the analysis are the Inhomogeneous Wave Correlation method, the Prony method and the time-of-flight method. The first two methods have been implemented on the same data set by exploiting measurement along a dense mesh of scanning points in the target direction with a 3D-Laser Doppler Vibrometer and using a piezoelectric transducer for generating waves. The third method requires a different data set, gathered using an instrumented hammer as a source and four accelerometers as receivers and uses the cross-spectral density function for calcula. The results show a remarkable match between the first two methods, while the third method requires an accurate choice of the time window to provide accurate results. Nevertheless, the third method provides a quick and low-cost alternative to the use of the laser vibrometer that can be more suitable for in situ measurements.
	Using near-field acoustic measurements to characterise mechanical and acoustic properties of lightweight building structures Andrea Santoni, Patrizio Fausti, Paolo Bonfiglio [Abstract] The evaluation of the vibrational field on a certain surface has great importance in noise control engineering applications. It can be helpful, both for noise reduction purposes, and diagnostic purposes, in product optimisation, or in order to characterise the mechanical properties of the vibrating structure. Accelerometers are still the most used sensors to measure vibration. However, contactless transducers, such as laser scanning vibrometers, have been widely used recently, presenting several advantages. Near-field acoustic holography may represent a valid alternative to LSV, in order to reconstruct the vibrational field on a surface from sound pressure measurements performed with an array of microphones. In this study, the vibrational field induced by an electro- dynamic shaker on a plywood panel was reconstructed from the sound pressure measured close to the vibrating surface. The test rig used allows to scan the sound pressure over the vibrating surface on a grid of points, using an array of microphones. The reconstructed panel’s dynamic response was used to evaluate its elastic and acoustic properties. The accuracy of this experimental approach was assessed by comparing the results with the ones obtained using miniature accelerometers. The advantages of the method are discussed together with drawbacks and limitations.
	Structural reverberation time measurements on WOODSOL prototype Simone Conta, Ulrich Simon, Anders Homb [Abstract] The WOODSOL research project aims at developing urban timber buildings based on moment resisting frames. One of the key aspects when dealing with the sound insulation in the building system is the flanking sound transmission across the floor elements and through the special connecting elements of the floor to the columns. We build a system prototype to be able to investigate experimentally the vibroacoustic behavior of the floor elements and the vibration transmission to the columns considering in particular impact excitation. In this paper, we focus on the structural reverberation time, which is a key parameter to determine calculation quantities such as e.g. Kij required by the ISO12354. The Kij-index is necessary to estimate flanking sound transmission and therefore a key input to SEA based calculation models. We present the measurement setup used at the Woodsol prototype and the measurement results obtained following the ISO 10848-1. Results will be obtained from both impact hammer excitation and shaker excitation with swept sine. We discuss the challenges we encountered and present the further steps in our analysis strategy.
	Bending wave based characterization of viscoelastic materials Max Miller, Sadeq Malakooti, Tahereh Taghvaee, Ning Xiang, Hongbing Lu, Nicholas Leventis [Abstract] Obtaining dynamic mechanical properties of viscoelastic solid materials can be challenging at acoustic frequencies. Commercially available characterization equipment is typically limited to about 500 Hz. Time- temperature superposition extrapolation to higher frequencies is indirect and time consuming. Resonance methods do not yield broadband data and fail for materials with high loss moduli. A new relatively simple characterization method, relying on bending wave excitation and laser Doppler vibrometer measurement, is proposed. Complex elastic moduli and phase speeds are now attainable to frequencies beyond 10 kHz with high reproducibility. The materials under test are an array of polyurea aerogels with fixed chemical composition but different nanomorphologies. Low cost and ease of synthesis add appeal to materials already noted for their sound transmission loss performance.
	An indirect method for the measurement of impact sound insulation Simon Bailhache, Stéphanie Colin, Michel Villot [Abstract] This paper describes a way of measuring impact sound insulation for heavy structures without using the standard tapping machine. Instead, another source of known contact force - e.g. an impact hammer or electrodynamic shaker - is connected to the building element under test. The sound pressure level is measured in the receiving room and then corrected from the difference in force level between the standard tapping machine and the substitution source to obtain the impact sound pressure level. An application of this indirect method is the laboratory determination of the impact sound performance of walls, which can be used as an input for prediction of impact noise of isolated heavy stairs as well as of structure-borne noise from service equipment in situ; the standard tapping machine is indeed not suitable for exciting vertical building elements. In this paper, the indirect method is described, then applied to a floor element and the results are compared to the impact sound pressure level directly determined using the standard tapping machine on the same floor. Several measurements are performed using different numbers of source and microphone positions. The corresponding results are presented and discussed.
	Continuing Prediction of Heavy/Hard Impacts on Resilient Sports Floors in Existing Buildings Matthew Golden, Faiz Musafere [Abstract] Previously the authors have presented a method to accurately predict the vibration performance of resilient sports flooring in existing buildings. The method includes the measurement of the force pulse injected into a building by a heavy weight drop onto resilient sports flooring. This force pulse is combined with an in-situ transfer function to predict in-situ performance. Extensions to the model include predictions over greater distances and multiple structure types. Further work extends the prediction model to include one-third-octave-band sound pressure levels. Finally, an alternative one-third-octave-band methodology will be shown and compared to one-third-octave-band prediction methods suggested by others.
	Structure-Borne Sound Isolation of Acoustic Test Chambers: In-Situ Validation Abhay Rajmane [Abstract] Acoustic test chambers are controlled environment for diverse acoustic and vibration measurements. They are broadly classified as Anechoic Chambers and Reverberation chambers. The background noise level inside an acoustic test chamber is important factor during measurements of silent specimens and sensitive equipment. Often the acoustic test chambers are built in noisy industrial halls. To achieve low background noise level, it is inevitable to structurally isolate the acoustic test chamber from surrounding structure. This is performed using vibration isolators e.g. G+H MAFUND Rubber Sheets, VIBREX Spring Strips or Metallic Springs. These System minimize the structure-borne sound entering in the floor of acoustic test chamber, thereby minimize the radiated airborne sound in the chamber. The elastic isolation is designed in such a way that the natural frequency is sufficiently below the cut-off frequency of acoustic test chamber. To Validate the elastic isolation, the natural frequencies are experimentally measured on floor of test chamber. Using the experimental modal analysis, the natural frequencies and mode shapes of the test chambers are extracted. It is validated that the measured natural frequency practically coincides with designed fundamental frequency. The experimental results from several acoustic test chambers are presented in this work.
	Qualification of an Anechoic Chamber Anthony Nash [Abstract] An anechoic chamber is typically used for acoustical experiments requiring a free-field environment. The free-field properties of the chamber can be determined by using a so-called ”divergence loss” (i.e., a draw-a-way) test to demonstrate that sound pressure decreases with increasing distance in accordance with the inverse square law. For the qualification of free-field environments, the principal governing standard is ISO 26101:2012. This standard prescribes protocols suitable for qualifying both hemi- and fully anechoic chambers including tolerances for deviations from the inverse square law. Either random noise or pure tones can be used for the ISO divergence loss test method, depending on the intended application of the chamber. This paper describes an ISO qualification test for a medium-sized anechoic chamber with an internal volume of 50 cubic meters. The chamber was equipped with a partly-open floor grate and furnished with foam wedges having questionable sound-absorptive performance. The chamber could satisfy the ISO tolerances when using random noise but failed to qualify when using pure tones. The paper discusses the measurements and the likely causes of the discrepancy.
	Sound Absorption Measurement Method using Ensemble Averaging Technique: A Robust Method for Surface Impedance Including in-situ Applications Otsuru Toru, Reiji Tomiku, Siwat Lawanwadeekul, Daiki Matsuoka, Ryota Yoshimoto [Abstract] Sound absorption characteristics of materials are of importance for controlling the quality of sound fields as well as of materials. In our previous paper, we proposed the concept and measurement method for surface normal impedance and sound absorption coefficient at random incidence condition, namely, ensemble averaged surface normal impedance and sound absorption coefficient. We named the measurement method as EA method, for short. We also showed the theoretical validation using the boundary element method and experimental confirmations on the effectiveness of EA method in successive several papers. In this paper, the outline of EA method is summarized, first. Then, reproducibility of the results are examined on measurements conducted both in reverberation rooms and in-situ. All the measurements were conducted by EA method with a pressure-velocity sensor (Microflown PU-sensor) which was calibrated on site in an acoustic tube. In-situ measurements were conducted at office room, living room, cafeteria and outside terrace. The results reveal that rather robust absorption coefficient values with the uncertainties less than 0.04 ∼ 0.05 are obtained regardless of measurement places and that EA method has broad adaptability to sound absorption measurements at various conditions.
	Measurement of Oblique-Incidence Absorption Coefficients of Various Types of Absorbing Materials in a Thin Chamber Tetsuya Sakuma, Naohisa Inoue, Yuta Sakayoshi [Abstract] A novel measurement system for oblique-incidence absorption coefficients has been developed using a thin rectangular chamber and two arrays of microphones. In this method, a series of incident and reflected plane waves at corresponding angles are decomposed based on a propagation mode expansion technique. Accordingly, absorption coefficients are determined at discrete incidence angles, of which number increases with frequency. First, a test measurement for glasswool was done in comparison with theoretical estimates. As a result, the measurement system was basically validated, but abnormal values were occasionally observed at singular combinations of frequency and incidence angle. The faults theoretically occur under two conditions: the interval of microphone arrays coincides with half wavelength in incidence angle, and some modes are not generated due to loudspeaker position mismatch. Therefore, a post-process was additionally introduced to omit abnormal values systematically. Then, measurements were done for various types of absorbing materials, such as urethane foam, rockwool board, perforated panel and honeycomb resonator, and their incidence angle dependence of absorption coefficients was examined.
	Advanced Investigation Using The EApu Method on the Effect of Quantitation and Particle Size of Charcoal in Clay Bricks on Sound Absorption Coefficient Siwat Lawanwadeekul, Otsuru Toru, Reiji Tomiku, Hiroyasu Nishiguchi [Abstract] Due to rapid urbanization, noise pollution becomes the environmental problems in Asia and their harmful effects on people’s health are undeniable. Reducing noise by absorption material is a good alternative. Bricks are one of the most common construction supplies in Asia because of the cost. Therefore, this study aims to apply the authors’ EApu method (sound absorption characteristics measurement method using ensemble averaging technique and a pressure-velocity sensor) to clay bricks to clarify the relationship between the sound absorption coefficient and the amount and size of an additive. The measurement was conducted in a reverberation room following the procedure presented in our previous papers. Measured bricks have four types of porous that were designed by selecting from different sizes of charcoal that was added into the specimens in the amounts of 0-30%wt. Next, the samples were fired at 900-1100 C and tested for sound absorption coefficient. Lastly, the microstructure analysis was conducted. The results indicated that the larger size and higher percentage of charcoal added in fired clay bricks presented the excellent absorption coefficient. Such a detailed investigation can be performed by the EApu method, and the EApu method helps material designing process become more accessible to control sound absorption.
	A Measurement Technique of Sound Absorption Coefficient and Impedance Using an Impedance Tube and Two Cardioid Microphones Kazuma Hoshi, Toshiki Hanyu [Abstract] Sound pressure and velocity can be measured using two cardioid microphones which are located at same position and pointed to opposite direction each other. This technique is called C-C method. Applying this technique, we tried to measure acoustic impedance of glass-wool 32 kg bulk density 50 mm thickness using an impedance tube and two cardioid microphones. The result is compared with a measurement result of two (omnidirectional) microphones technique. As results, it is clarified that surface impedance of material can be measured by two cardioid microphones over all frequency range in limitation of the impedance tube.
	A new technique for measuring sound particle velocity and sound pressure using face-to-face cardioid microphones Toshiki Hanyu, Kazuma Hoshi [Abstract] Generally sound particle velocity can be measured by using a particle velocity sensor directly or by two sound pressure type microphones. Impedances and absorption coefficients of materials can be measured by these sensors. We have proposed a new technique for measuring sound particle velocity and sound pressure using face-to-face cardioid microphones, which is called the C-C method. A cardioid microphone responds to both sound pressure and particle velocity. Directivity of the cardioid microphone is formed by a combination of both sound pressure and particle velocity. In our new technique, the sound pressure and particle velocity can be obtained simply by summation and subtraction of responses of the face-to-face cardioid microphones. However, errors of measured sound pressure and sound particle velocity occur due to differences of directivities of the cardioid microphones. Therefor we have developed a calibration method of the cardioid microphones. By this calibration, accurate sound pressure and particle velocity can be measured even by using any kinds of cardioid microphones. We have conducted measurements in order to verify the new technique. Results of the measurements supported that the new technique can be used for measuring sound pressure and sound particle velocity accurately. Advantages of this technique are also discussed.
	Airflow Resistance Measurements between Room Temperature and 800 C Thamasha Samarasinghe, Carl Hopkins, Gary Seiffert, Jilly Knox [Abstract] The acoustic properties of porous materials are usually measured at room temperature but in many industries, porous materials are required to provide sound absorption in high temperature applications such as in Heat Recovery Steam Generators (HRSG) applications, automobile silencers and aero engine liners. In this paper, measurements have been made of the airflow resistance of three different densities of Alkaline Earth Silicate (AES) fibrous materials. Measurements at room temperature to ISO 9053 and in a bespoke test rig at temperatures up to 800∘C were made using a kiln to achieve the high temperatures. The high temperature tests on these particular materials indicated that it might be possible to assume no significant effect of temperature on specific airflow resistance between 20 and 100∘C but not at higher temperatures. Above 600∘C, the effect of crystallisation and the change in material thickness meant that specific airflow resistance (rather than airflow resistivity) was used in regression analysis to identify empirical relationships for the three different density materials at different temperatures.

Organization:

D. Bard-Hagberg, Lund University A. Homb, SINTEF Building & Infrastructure

	Reducing ground-borne noise due to railways. Part I: assessing the problem Catherine Guigou-Carter, Guillaume Coquel, Philippe Jean, Alexandre Jolibois [Abstract] The vibration field along a railway line was measured and modeled to predict its impact on a future real estate project. Indeed, vibration levels due to rolling stocks evaluated near the tracks are usually evaluated on the vertical axis and at one or a few positions, supposed to be representative of the entire vibration field around the tracks. When the project size is rather extended along the tracks, this assumption might be incorrect, especially if the project site is located near a train station, where stresses in the tracks can be very different depending on the direction of the train due to acceleration or deceleration. This paper presents vibration measurements performed in the different directions before buildings construction along the tracks. Furthermore, the attenuation of ground vibration as a function of distance from track is used to evaluate the loss factor in the ground based on a comparison with a 2.5D BEM model of the ground including the different layers. The model is then used to identify equivalent forces in the different directions associated to train passage to obtain similar vibration transmission behavior in the ground. The mitigation measures are discussed in a companion paper.
	Reducing ground borne noise due to railways. Part II : mitigation measures Philippe Jean, Catherine Guigou-Carter, Alexandre Jolibois [Abstract] The construction of new buildings very close to existing tracks with heavy commuting traffic requires careful design. Such projects must be accompanied with noise reduction measures. The practical case here considered has been analysed by first assessing the existing situation prior to construction (type of soil and traffic: results exposed in a companion paper). Here we focus on the mitigation measures which have been tested and the expected corresponding vibration and ground-borne noise levels inside the buildings. A 2.5D FEM/BEM commercial software (Mefissto), developed by CSTB, has been employed in what is a very challenging numerical context: multi-layered ground, large multi-levels buildings, large frequency range (10-200 Hz). In every situation the use of an original 2.75D approach leads to the noise levels inside the building. As the tracks cannot be modified the solutions tested consist in inserting different types of barriers in the ground or in inserting spring-like materials in the lower part of the structure. Results are compared to target values in terms of vibration and ground-borne noise levels selected. In order to limit dwellings occupants’ annoyance. Satisfactory results can only be achieved with the use of the spring-like solution.
	Correction of Sound Pressure Levels Calculated for Small Rooms Using Diffuse-Field Theory Kiyoshi Masuda, Hikari Tanaka [Abstract] The most frequently used method for predicting sound pressure levels arising in a room as a result of transmitted sound or floor impact sound from adjacent rooms is an acoustic energy calculation assuming a diffuse sound field. If the dimensions of the room under investigation are sufficiently large, sound pressure levels can be predicted accurately by this method. However, especially at low frequencies, the wavelength of the sound may be nearly equal to or exceed the room dimensions, leading to reduced predictive accuracy. Although the Waterhouse correction is typically used to correct for this error, its reliance on statistical energy considerations means that it cannot take into account the response to the frequency range below the first-order mode and to frequencies between adjacent lower-order modes. In this work, a new correction method is derived from the relationship between the sound pressure level calculated using wave theory and the sound pressure level using diffuse-field theory. By comparing calculated results obtained using the new method with both actual measurements and FEM calculations, its accuracy is examined.
	Using realistic test signals to evaluate existing structures for low frequency sound transmission from clubs, live music venues, discos, and exercise facilities David Sage Woolworth [Abstract] Low frequency sound used in entertainment (and sometimes exercise) facilities is difficult to reproduce in standard testing procedures for existing facilities such as ANSI E336 and ISO 10848-1, as the test cutoff frequency is well above sounds generated in-situ for these facilities, and some methods or loudspeakers cannot reproduce the sound level of low frequency that would be typical for everyday operation for these facilities This paper proposes using alternate test signals and methods that are appropriate to facility programming that must be isolated from neighboring structures. It addresses ”worst case” low frequency source characterization, sound levels that incite vibration of finish materials that are transferred through the structure, and metrics used to evaluate receiving spaces.
	Experimental Study on Low-Frequency Averaging of Indoor Sound Pressure Level in Façade Sound Insulation Measurement Jinyu Liu, Naohisa Inoue, Tetsuya Sakuma [Abstract] In the ISO 16283 series for field measurement of sound insulation, a low-frequency procedure is specified for determining indoor average sound pressure level, which is the so-called corner method. In the procedure, additional measurements are required in the corners in addition to the default measurements in the central zone, and the indoor average level is corrected with the highest level in the corners. However, this procedure is empirically proposed, and its validity is not fully examined for various cases. In this paper, façade sound insulation measurements are performed in the low frequency range for a mock wooden house and a 1/4-scale model house. In the rooms with the volume of about 20 m3, sound pressure levels are measured at multiple grid points including corners. Regarding the average level for all grid points as a reference, the ISO procedure and other possible averaging schemes are examined in the aspects of bias and deviation due to measurement point selection.
	Additional sound insulation panels ZIPS-experience of 20 years of application Alexander Boganik, Anatoly Livshits [Abstract] In apartment buildings and offices is often the task of increasing the sound insulation of walls and ceilings from airborne noise. Traditional methods of solving this problem, gypsum plating on metal or wood frame - have low efficiency. This is, in particular, due to transmission sound vibrations through the frame and the presence of a large number of wall lining elements that must be properly performed during installation. The frameless panels of additional sound insulation (ZIPS) were developed in 1999 in Russia. They have high rates of additional sound insulation (9...18 dB) in depending from thickness of 43...120 mm. This system was patented and is still widely used. In addition, the system provides the necessary strength and ease of installation. Over 20 years, the system has undergone several upgrades. The article describes the physical principle of the ZIPS-system, the data of laboratory and full-scale acoustic measurements, latest modification of the system with adjustable mounts.
	Noise from plants systems and Building Information Modeling: The Code Checking. Costantino Carlo Mastino, Roberto Baccoli, Andrea Frattolillo, Martino Marini [Abstract] The acoustic performance of buildings is increasingly a performance requirement required both in the construction of private buildings and public buildings. At international level, in recent years we have witnessed the revision and issue of new standards relating to the project phase (forecast calculation) of passive requirements based on the ISO 12354 series of standards. At the plant design level, the reference standard is represented by EN 12354-5 and ISO 16032 while in the post-operation verification phase the ISO 16032 and ISO 10052 standards are the reference for the measurements. One of the fundamental aspects to achieve high performance is to reach a level of design study that is increasingly detailed and shareable. Building information modeling, in addition to being an obligation for some types of work, is increasingly used by designers and construction companies, to have a better knowledge of the case on particular and methods of execution. The present work presents an overview of procedures that exploit the information of the BIM model for the predictive calculation of the acoustic performances of the systems, the verification of the legislative limits and the integration of the instrumental data recorded post-operam. The procedures were finally tested on a case study.

Organization:

F. Martellotta, DICAR-Politecnico di Bari V. Wittstock, Physikalisch-Technische Bundesanstalt

	Experimental characterisation of absorbing materials made from renewables Volker Wittstock, Daniel Sgriess [Abstract] The amount of ecological implications caused by the whole building process is gaining more and more attention. One measure to reduce ecological implications may be the use of building material made from renewables. To investigate this possibility, a joint research project is under way in Germany, which is focused on insulating material made from renewables. Besides fire protection, heat insulation, moisture proofing, emissions and sustainability assessment, also sound insulation is investigated. PTB contributes to this project by experimentally characterising insulating materials made from renewables. For this, existing measurement methods have been reviewed and adapted to the special needs of the project. The contribution is focused on the measurement of airflow resistivity by the alternating flow method and measurements in a transmission tube. Special attention is given to the influence of the material orientation and to the uncertainty of the measurement results.
	Investigations regarding the Influence of Static Load and Airflow Resistance on the Measurement of Dynamic Stiffness Heinrich Bietz, Martin Schmelzer, Volker Wittstock, Spyros Brezas [Abstract] Within the scope of a research project regarding the physical properties of insulation materials made from renewables, one of PTB’s tasks is to determine the dynamic stiffness of such materials used as elastic layers. The common measurement procedure is described in DIN EN 29052-1. As a closer examination of this standard bears some open questions, several investigations have been conducted recently. This presentation focuses on two aspects: The first one is the influence of the static preload. This might be relevant for insulation materials used for thermal insulation composite systems, where, compared to floating floors, the static preload is significantly smaller. As the standardised measurement procedure is not suited for small static preloads, an alternative measurement set-up was designed to investigate this topic. The second aspect is the effect of airflow resistance on the resulting dynamic stiffness. The measurement setup as devised in the standard does not necessarily reflect the boundary conditions the material is exposed to when it is mounted in situ. The standard describes a procedure to handle the influence of airflow resistance, but this is partially incomplete and not applicable for some materials. Some theoretical considerations and measurement results regarding this topic will also be presented.
	Insulation materials made of renewable raw materials for the sound insulation prognosis of building components Simon Mecking, Andreas Rabold, Anton Huber [Abstract] In order to increase the application possibilities in the building sector for insulating materials made from renewable raw materials, a project was initiated by the Fachagentur Nachwachsende Rohstoffe e.V. which shows the different requirements for these insulating materials and provides planning and verification options. In the field of sound insulation, building component catalogues for these insulating materials are expanded, test methods for the material properties of the insulating materials are defined and calculation models are further developed. The airborne and impact sound transmissions of the components are predicted in order to investigate the influence of the different insulating materials from renewable raw materials in comparison to conventional insulating materials. Component measurements indicate that a characterization of the cavity-insulation alone by the flow resistance is insufficient. Using a semianalytical calculation approach with an equivalent fluid, further insulation material properties can be taken into account in the prognosis of double wall elements. The input data for the prognosis calculation are determined by measurements of the transmission properties and compared with results of other calculation models based on the flow resistance. By validating the first prediction results with measurements, it should be shown to what extent the conventional measured variables are sufficient.

Organization:

F. Martellotta, DICAR-Politecnico di Bari V. Wittstock, Physikalisch-Technische Bundesanstalt

	Preliminary investigation on the acoustic properties of absorbers made of recycled textile fibers Chiara Rubino, María Angeles Bonet-Aracil, Stefania Liuzzi, Francesco Martellotta [Abstract] In recent years, wastes reduction and their conversion into new resources are becoming a major requirement for our society. A considerable interest is focused on the use of recycled textile materials in building products and on their potential application as absorbing acoustic material. In developing innovative products it is not only important to recycle, but also to manufacture devices with recycled content. The challenge is the use of textile waste fibers and nontoxic binder to obtain composite materials with less social, economic and environmental impact. In this paper 100% merino wool wastes bounded with chitosan are investigated to obtain sound absorbing materials, particularly for the acoustic refurbishment of the open- office spaces. The use of chitosan instead of synthetic and plastic elements represents a valid solution to obtain bio- materials for optimizing buildings environmental sustainability. The acoustic performance of innovative composite materials were investigated. Measurements of normal incidence sound absorption coefficient were carried out by means of a standing wave tube, according to ISO standard 10534-2. Experimental results were very promising, showing absorption coefficient higher than 0.5 at frequencies above 500 Hz. This proved that it could be possible to produce sustainable alternative to traditional commercial synthetic products.
	Characterization of woven fabrics for development of micro perforated panel absorbers . Gunawan, Iwan Prasetiyo, B Yuliarto, K Anwar, D R Andhika [Abstract] Woven fabrics have a potential to be developed as sound absorber system because the presence of micro perforation. By introducing air cavity backing the woven fabrics up, such system can undergo Helmholtz resonator mechanism as found in micro-perforated panel absorber. The micro perforations in a woven fabric are formed by yarn in x and y direction. The perforations can create a viscos-inertial effect when interact with sound fields that is useful as a basis for sound absorption mechanism. In this study, we are focused to investigate the relationship of woven fabric material and manufacturing technique to sound absorption characteristics. Natural and synthetic fibers-based yarn are used to produce woven fabrics. For investigation, the geometrical and physical properties of the yarns and fabrics are characterized. The geometrical properties are characterized using Scanning Electron Microscope (SEM) while the physical properties are measured using textiles testing devices. The sound absorption coefficients are evaluated by impedance tube using transfer function method. From this study, it can be concluded that the sound absorption coefficients are affected by material and manufacturing technique properties. It is expected that the results can be beneficial for developing design procedure of development woven fabric based micro perforated panel absorber system.
	Sound Absorption Characteristics of a Back-Perforated Honeycomb Panel with an Air-Layer Kyungsoo Kong, Tetsuya Sakuma, Naohisa Inoue [Abstract] A honeycomb panel has a high potential in noise control engineering as well as in any other engineering fields. This paper demonstrates that a back-perforated honeycomb panel (BPHP) placed with a back air-layer exhibits unique absorption characteristics composed of the combination of panel- and resonator-type mechanisms. A BPHP is theoretically modeled as a two-mass system composed of panel and resonator’s neck masses, and multiple air stiffnesses. First, resonance mechanisms are clarified from the model. It is revealed that the panel-type resonance can arise from the panel mass and air stiffness corresponding to the total thickness of the panel and air-layer, even if a BPHP is placed considerably closely to the back wall. Next, normal- and random-incidence absorption coefficients are calculated with changing several parameters such as perforation diameters, hole pitches, panel thickness, back air- layer thickness and so on. Furthermore, a supplementary numerical calculation confirms the validity of the proposed theoretical model. Through the case studies, several charts are arranged for effective design of BPHPs as low-frequency sound absorber. Finally, absorption performances of BPHPs are measured by the reverberation room method, and the results fairly agree with those calculated by the proposed model.
	Architectural Acoustic Design using absorption materials: the case study 'Snooze Panel' Livio Mazzarella, Maria Cairoli [Abstract] The Snooze® sound-absorbing panel is the result of the necessity to characterize an architectural space such as offices, corridors, classrooms, hospitals while guaranteeing the required acoustic characteristics using renewable materials. An experimental case study is proposed showing how different chooses in panel assembly arrangement are affecting its acoustic performance. The proposed procedure allows evaluating the acoustic properties of a panel assembly, specifically the rating of sound absorption according to the EN ISO 11654 i.e. the weighted sound absorption coefficient, αw, considering the intrinsic panel characteristics and the different types of installation with variable distance from the wall surface behind the panel. While the material characteristics can be assessed in laboratory using appropriate instruments as explained, the acoustic behavior of the panel due to the geometry and installation solutions are evaluated in actual assembly in reverberation chamber according to the standard EN ISO 354.
	Analysis of environmental and seasonal effects on sound absorption by green wall systems Emmanuel Attal, Nicolas Côté, Takafumi Shimizu, Bertrand Dubus [Abstract] Buildings and roads are made of acoustically rigid materials which reflect sound emitted from many sound sources and contribute strongly to acoustic environment In that context, green roofs and/or walls used for building envelope greening were found to be efficient solutions for acoustic protection in cities when applied in a significant urban scale. Green systems (roofs and walls) are usually composed of plants, soils and/or air gaps. Below 1000 Hz, sound attenuation is low in plants and air but much higher in soils. However, sound absorption by green systems vary strongly with the arrangement of plants, air and soil and efficient broadband absorption is only obtained when the arrangement is optimized. In this work, the effect of green wall arrangment as well as environmental (moisture content) and seasonal (leaf density) effects on absorption coefficient are studied. Effective properties of plants and soils are measured in an impedance tube using three-microphones-two loads method between 100 and 1000 Hz under controlled moisture content for soils and leaf density for plants. Sound absorption coefficient at normal incidence is then calculated using matrix manipulation. Variation of absorption coefficient with moisture and leaf density is provided for various green wall systems.

Organization:

K. Larsson, RISE Research Institutes of Sweden L. Barbaresi, University of Bologna

	Facade Sound Insulation as Protection to Outdoor Noise Chiara Scrosati, Fabio Scamoni, Michele Depalma, Matteo Ghellere [Abstract] In recent years the acoustic qualification of buildings has gained more and more importance at international level. One of the important elements for determining the sound quality of houses is the insulation against external noise. In fact, many issues related to excessive noise in dwellings arise from their inclusion in noisy environments, or from their original surrounding environment, which has grown from silent to very noisy over the years. The main strategy to limit the noise inside the buildings is the façade insulation improvement. Regulatory requirements and classification schemes in Europe present a high degree of diversity. In particular, the approach to the façade insulation acoustic requirement adopted in the various countries is very diversified also because closely related to the external noise climate. The main issues related to the façade sound insulation are investigated with particular regard to the choice of descriptors, to the evaluation methods and their reproducibility, which is the less easy to evaluate.
	Measurements and prediction of sound insulation of innovative ventilated façade solutions Francesca Di Nocco, Luca Barbaresi, Federica Morandi, Massimo Garai [Abstract] This work is part of ”Progetto Involucro”, a research project funded by the Emilia-Romagna Region in the frame of POR-FESR 2014-2020. It aims at testing innovative ventilated façade solutions for improving buildings’ energy and acoustic performance. Two identical test buildings (prototypes) are built in real scale on the same site and different ventilated façade solutions are installed on the South facing walls. In order to evaluate the performance of different alternatives, a benchmark solution is installed on the first prototype, while all the other innovative solutions are installed on the second one and they all are tested under the same environmental conditions. Sound insulation measurements are performed on site with the aim to investigate the contribution of different rainscreens in relation to their fastening system and individual component elements such as natural ventilation grills, open joints and potential openings on the external side. Furthermore, a measurement campaign focused on flanking transmission is conducted to evaluate sound transmission across the façade junctions for each innovative solution installed. This paper shows the preliminary results of the measurement campaign. Data obtained from on-site measurements are commented and discussed in relation to the values calculated using the model proposed in the ISO 12354-3 standard.
	Auralization of the Coincidence Effect on Façade Sound Insulation Dag Glebe, Krister Larsson [Abstract] If a façade contains windows, their sound insulating properties will often govern the insulating performance of the façade. When presenting the façade’s insulating effect on the in-door sound environment, a natural approach is to auralize a realistic example of traffic noise that has been acoustically filtered by a window. One approach is to use results from measurements of the window’s sound insulating properties according to ISO 10140-1. The third octave band sound reduction indices can then be used to filter a traffic noise sound file. However, some important properties are not covered by this approach but should be considered. One example is the coincidence effect, which is only represented as an integration of noise incident at all angles, due to the diffuse field, and thus only contribute to the amplitude. However, the relationship between incidence angle and frequency may be important for the perception of traffic noise, particularly at single vehicle passages. Therefore, two auralization approaches were evaluated in a listening test with respect to how they were perceived: the simple amplitude-based approach with data from standard measurements, and an approach employing the transfer function of different incidence angles.
	Designing a Better Plenum Window or Balcony for Higher Noise Reduction Against Outdoor Noise Liu Yee Louisa Cheung, David Bk Yeung, Ching Chan, Calvin Chiu [Abstract] Plenum windows have been used as innovative measures to reduce traffic noise in the past decade. Nowadays, they are also used in mitigating more serve traffic noise, railway noise or even fixed noise which require different noise reduction characteristic and enhanced performance. To enhance the noise reduction performance of the plenum windows and balconies, different kinds of absorption treatment like micro-perforated absorbers and perforated absorption panels had already been used and tested. They could provide around 1dB of improvement. Meanwhile, window designers are finding alternative ways to further enhance the performance, such as by changing the physical design and dimensions. The most common changes are to increase the overlapping length of the plenum of the window or by adding a solid parapet in front of the plenum balcony door. Numerous designs of windows and balconies with different dimensions have been tested, with different overlapping length, different acoustic treatment and even outer window or balcony door design. The tests were done in full scaled mock up using linear array of loud speakers mincing traffic noise source. The effects in noise reduction of implementing these design changes to plenum windows/balconies are being studied by comparing the measurement results in this paper.
	Sound Insulation of Fenestration Systems: A Comprehensive Web-Based Simulation Program and Validation Tejav Deganyar, Fangliang Chen, Yihe Huang, Rebeccah Grotemeyer [Abstract] Aluminum framed fenestration systems are becoming leading choice for modern buildings, as they provide a better aesthetic appearance, higher degree of weathering reliability, and quicker installation while maintain more transparency and environmental efficiency. Because of the introduced complexity of the systems, however, estimation of sound insulation of such systems for preliminary design or analysis become much more time consuming and difficult. Currently, no straightforward solution to fulfil such purpose is available, and a convenient and reliable tool is highly desirable to provide a quick estimation of the sound insulation performance of fenestration systems. Bearing this burden, the engineer team led by the authors conducted extensive studies on various fenestration systems consisting of different glazing and aluminum frames. Based on our research outcomes, a comprehensive web-based simulation program, so-called Digital Acoustic Lab (DAL), has been developed recently. This program is able to provide quick analysis and accurate prediction of sound insulation of window, door and façade systems consists of various aluminum frames and glazing infills. A series of laboratory tests were conducted to validate the predictions by the DAL. The goal of this program is to provide designers, engineers and architects an effective and economic tool to facilitate the design.
	Adaptive acoustic comfort: Assessing noise with provisions for ventilation and overheating in dwellings Jack Harvie-Clark, Anthony Chilton, Nick Conlan, David Trew [Abstract] In the design of residential developments in the UK, it has been common practice for façade sound insulation (to protect against outdoor noise) to be considered separately from the provisions for ventilation and for mitigating overheating. This fragmented approach has led to different designers making different, incompatible assumptions about the internal environmental quality conditions: the acoustic designer assumes that windows are closed to control external noise ingress, while the mechanical designer assumes that windows are open for ventilation or mitigating overheating, for example. This leaves occupants with a choice between reasonable noise levels or thermal comfort, but not both. This problem is exacerbated by increased overheating risk in modern buildings and future climate scenarios. In response to this issue, the Association of Noise Consultants has produced the Acoustics Ventilation and Overheating Residential Design Guide - the ”AVO Guide”. The AVO guide recommends an approach to acoustic assessment that takes regard of the interdependence of provisions for external noise ingress, ventilation and overheating. This paper describes the context, background and content of the AVO Guide. There are references to attenuated options for ventilative cooling to help designers avoid simply specifying mechanical cooling.
	Requirements for the façade sound insulation for different types of outdoor noise Steffen Körper [Abstract] Buildings are exposed to a wide variety of external noise situations. The time course of the sound pressure level and the frequency response of external noise can vary greatly between the vicinity of roads compared to the vicinity of airports or railways. The sound insulation of facades protects the indoor space and the residents against external noise day and night. The German standard DIN 4109 ”Sound insulation in building construction” contains the minimum requirements for sound insulation for all new buildings in Germany. All requirements are based on a single-number-value, the equivalent sound pressure level, without adequate respecting to the particularities of the frequency response and time structure of a sound source. The presentation shows the German standard values in building acoustics and the subject of external noise. The results of measurements of different sound sources will be shown. A new approach and an outlook for further possibilities to protect inhabitants appropriately to the noise situation will be discussed.
	Ventilative cooling in noisy environments: practical options for the UK Nick Conlan, Jack Harvie-Clark [Abstract] It is only very recently that a proper consideration of the internal acoustic environment when mitigating overheating has started to be taken seriously in the UK. Previously this issue has not been addressed by the various regulatory systems, and developers have been reluctant to pay for what they have seen as an unnecessary consideration. The Association of Noise Consultants has produced the criteria and guidance in its Acoustics, Ventilation & Overheating: Residential Design Guide. At this crucial stage in the establishment of new approaches for the construction industry, we are setting out ”rules of thumb” with typical options for ventilative cooling in varying external noise environments. We are developing solutions in collaboration with mechanical engineers and product manufacturers that are suitable for UK building typologies. Vents in the external facade are found to present the easiest adoption into current building designs. As acoustic attenuation typically involves airflow resistance, mechanical fans are required to ensure ventilation rates are achieved in noisier environments or where the apartment has single- sided ventilation opportunities.
	Innovative Noise Mitigation Measure- Baffle Type Acoustic Window Ching Chan, Liu Yee Louisa Cheung, David Bk Yeung, Calvin Chiu, Billy Fan [Abstract] Hong Kong is one of the densely populated area in the world with residential sites surrounded by traffic roads. Road traffic noise pollution problem becomes a major environmental issue for residents in Hong Kong. In an attempt to tackle road traffic noise impact on residents and ensure good quality of individual’s health and well-being, an innovative measure called Baffle Type Acoustic Window has been proposed, which designs to benefit residents from environmental noise planning and building ventilation’s aspect. It consists of two glazing: (i) an outer openable window and (ii) an inner sliding panel with air gap in between. The inner sliding panel is situated right behind the outer openable window. To investigate the acoustic performance of Baffle Type Acoustic Windows with different design parameters, a series of acoustic laboratory tests have been carried out in accordance to ISO Standard to discover the relationship between the key design parameters of the window (e.g. width of overlapping between sliding panel and the window system, size of outer window opening and acoustic material used etc.) and its corresponding acoustic performance.

Organization:

J.Y. Jeon, Hanyang University C. Hopkins, Acoustics Research Unit, University of Liverpool

	Development of a modified impact testing method for simultaneously evaluating multiple floor toppings Wayland Dong, John Loverde [Abstract] Materials such as floor coverings, screeds, and resilient matting are commonly applied on top of a floor-ceiling assembly to improve the impact noise insulation. Theory and experience show that the benefit of such materials increases with frequency, with minimal effect below the natural frequency of the resilient materials and the largest effect at high frequencies. This improvement in impact insulation is added to the base performance of the structural assembly. Floor design can involve multiple parameters (finish floor, screed thickness, resilient mat thickness) even from the same manufacturer or product line, and the number of permutations rises quickly. These permutations must also be tested on multiple structural assemblies. The authors are investigating a modified method where multiple top-side assemblies could be installed on the same structural assembly, allowing for a more efficient testing process. The expectation is that airborne and low-frequency impact insulation would be consistent since the base assembly is unchanged, but high-frequency impact insulation would provide the same result as a full-scale test. Data from the investigation is presented and analyzed to validate the model and document the uncertainties.
	Round robin for testing tapping machines Sylvia Stange-Kölling, Volker Wittstock [Abstract] From a metrological point of view, tapping machines are working standards for a defined vibrational excitation of building elements. To ensure a uniform excitation, tapping machines have to comply with several requirements according to the relevant ISO-standards. Whether a tapping machine is in conformity with these requirements is tested by different laboratories in Europe. Since the requirements are well defined but not the way the conformity has to be tested it is questionable whether different laboratories come to the same results for the same tapping machine. On that occasion, a round robin on tapping machine testing was organized by PTB. Two different tapping machines were tested in 5 different laboratories in Europe. The main quantities hammer curvature, time between hammer impacts, velocity of the hammer at the impact and hammer mass were measured by all laboratories. Results of these measurements are presented and possible future developments are discussed.
	New single-number quantities for evaluation of impact sound insulation Mikko Kylliäinen, Petra Karoliina Virjonen, Valtteri Hongisto [Abstract] The standardized single-number quantities for evaluation of impact sound insulation do not correlate especially well with the subjective judgment of living impact sound sources directed to the floors. The aim of this study was to find new single-number quantities better associated with the subjective annoyance caused by different impact sounds. The new single-number quantities were developed on the bases of experimental data of laboratory measurements of impact sound insulation of floors and a psychoacoustic experiment. Five studied impact sound types were walking with hard shoes, socks, and soft shoes, super ball bouncing, and chair moving. Basic requirement was that the new single-number quantities could be expressed as the sum of L’n,w or L’nT,w and new spectrum adaptation terms. Reference spectra for calculation of new spectrum adaptation terms for each sound type were derived by the means of a mathematical optimization method. In addition, an optimized reference spectrum based on all five sound types was derived.
	Efficient numerical prediction and experimental validation of impact sound radiation by timber joist floors Pengchao Wang, Cédric Van Hoorickx, Geert Lombaert, Edwin Reynders [Abstract] Achieving good impact sound insulation with a timber floor is challenging, given its low weight. Therefore, the acoustic performance for a timber floor should be considered at the planning stage, and a prediction tool is required to evaluate the impact sound insulation design. This work investigates the prediction of the impact sound radiated by a timber joist floor. The floor properties are determined by modal tests of the separate floor components. The impact sound radiated by the floor is subsequently computed in two steps. In the first step, the impact force of an ISO tapping machine is computed by considering both the global and the local mobilities of the floor. In the second step, the response of the point-excited floor is computed by the finite element method, and the sound pressure level in the underlying room is computed by a diffuse room model. Special attention is paid to the influence of boundary conditions on the computed eigenmodes of the floor; these are also compared against identified eigenmodes of the floors. Finally, the sound pressure levels and the single number rating results are computed, for excitation by a instrumented hammer and the ISO tapping machine. The computed results are experimentally validated.
	The preliminary study on subjective rating of different floors characterised by Ln,w+CI,50-2500 Vojtech Chmelík, Jakub Benklewski, Monika Rychtarikova, Dominik Kisić, Kristian Jambrošić, Marko Horvat, Herbert Muellner [Abstract] The problem of noise in dwellings is topic of large discussions nowadays. Not only airborne noise but also impact noise are responsible for decreased comfort of living. The subjective evaluation of impact noise sounds in dwellings are described in this paper. Listening tests based on method of adjustment (Scharf,1961) with statistically sufficient sample of tested subjects were performed in listening room in the laboratory of TGM where subjects compared performance of presented floors. Impact noise stimuli were filtered with various floor constructions with similar Ln,W values, sorted into two main groups: (i) light-weight (LW) and (ii) heavy- weight (HW). Comparisons of listening tests performance with calculated single number quantities for each construction were prepared in order to understand the perception of presented sounds by dwellers. The results from preliminary study are meant as a basis for adjusting of currently applied reference curve also with particular focus on the low frequency range.
	Perception of Combined Indoor Noise Sources in Lightweight Buildings Alessia Frescura, Pyoung Jik Lee [Abstract] Contrary to community noise field, the combined effects of multiple noise sources on annoyance have not been investigated in building acoustics. Therefore, the present study aims to quantify the total annoyance caused by indoor noise from the upstairs and next-door neighbours. Footstep sounds were recorded in a laboratory equipped with a timber joists floor in different configurations. Listening tests were then performed to assess the relationship between annoyance ratings and sound pressure levels in several scenarios. In the first part, the annoyance caused by individual impact and airborne noise was evaluated considering the sound insulation performances of different wall and floor structures. The participants were also asked to evaluate the total annoyance in the second part when impact and airborne sound sources were presented at the same time. This paper will show models of noise annoyance when impact and airborne sources are heard separately and simultaneously. The findings of this study would be helpful to understand how we perceive the combination of noise sources we are usually exposed in timber houses.
	Influence of suspended ceiling type of residential building on heavyweight floor impact sound Song Hansol, Jongkwan Ryu, Inho Kim, Yong Hee Kim [Abstract] This study investigated the influence of suspended ceiling type of residential building on heavyweight floor impact sound through floor impact sound measurements in test building using rubber ball. In general, gypsum board is suspended using hangers for the ceiling finishing in the Korean residential building, and air space between slab and the suspended ceiling thus occurs. Three types of the suspended ceiling (perforated, sound absorbent, and resonator ceiling) were tested to explore reduction of floor impact sound level in comparison with general gypsum board. Results showed that three suspended ceilings significantly reduced by 1∼4 dB in single number quantity for rating heavyweight floor impact sound insulation. It was also found that dominant frequency band of the reduction depended on the suspended ceiling types.

Organization:

A. Kohlrausch, Eindhoven University of Technology K. Bijsterveld

MATLAB-based simulation software as teaching aid for physical acoustics Jorge Petrosino, Lucas Fernando Landini, Georgina Alejandra Lizaso, Antonio Ian Kuri, Ianina Canalis [Abstract] This paper presents examples of simulated experiments that can be run in MATLAB or Octave as a demonstration for students, allowing them to explore and understand the basics of wave propagation. The use of the k-Wave toolbox is proposed. This open source add-on is capable of performing simple and efficient simulation of wave propagation in the time domain, as well as providing comprehensible real-time visualization of the process. It is mainly used to simulate ultrasonic waves in biological media; however, its reliability isn’t diminished in applications where lower frequencies are involved. In this work, 2D acoustic signals in the audible range are used to explore a variety of phenomena related to physical acoustics, such as reflection, diffusion, diffraction, absorption, resonance. The placement of the acoustic elements in the medium for each numerical experiment is extracted from a BMP image file depicting sources, sensors and solid materials, encoded in different colors. The proposed method allows users possessing elementary knowledge of MATLAB / Octave code to interact with the k-Wave toolbox in a beginner-friendly fashion. Vast control over the simulation conditions can be exercised by simply drawing, cutting, pasting, moving or recoloring elements of the image file.

Information Retrieval and Topic Modeling based on the ICA Proceedings Thore Oltersdorf [Abstract] Information retrieval is a key feature to get an idea on topics that are new to someone. It is a general purpose technique that any scientist is being educated in during his career. But times and techniques have changed rapidly in the past few years. Multiple approaches are possible now due to massively available open data. It is also a recursive process since often the linguistics and/or programming capabilities are missing but needs to be known before its usage can be started. Based on open source software tools approaches from NLP (natural language processing) will be demonstrated on a certain aspect of technical acoustics. This paper will introduce how ICA proceedings from 1953 up to 2016 are used to get an idea of acoustical aspects of technical systems used for HVACR. This is not the same as understanding and having the knowledge of each detail but it is key to gain it.

Organization:

J. Bruyninckx, Maastricht University V. Tkaczyk, Max Planck Institute for the History of Science / Humboldt-Universität zu Berlin

	From Noise Control to Intelligibility: a history of office acoustics Joeri Bruyninckx [Abstract] As historians have shown, at least since the early twentieth century, architectural acoustics has been concerned with dampening reverberant spaces into a clean, echoless acoustical ambience that came to represent the so-called ësoundscape of modernity’ (Thompson 1995). In this paper, I show how architectural acousticians actually developed a completely different approach to noise control, shifting from acoustic noise control to managing (psycho-)acoustic intelligibility. I will discuss this shift against the context of office planning. From the 1950s onwards, new office concepts (such as the open office) were embraced for their planning flexibility, efficiency and advantages to inter-employee communication. However, they also posed new problems, as employees complained of noise, reduced concentration and lack of privacy. Professional acousticians in the 1960s and 70s addressed these complaints by modelling ëprivacy’ as a problem of speech intelligibility. They did this, among others, by repurposing the Articulation Index, a wartime standard developed to evaluate telephonic communication, into an inverted measure of privacy. Drawing on archival research, on acoustical isolation manufacturer Owens-Corning Fiberglas and office design manufacturer Herman Miller Inc., I trace how sound masking prompted a series of new technologies, materials and approaches to ëcondition’ building acoustics–yet again radically changing the architectural soundscape.
	Sound & Science: Digital Histories: Navigations through the History of Acoustics Viktoria Tkaczyk [Abstract] This paper will present the database ”Sound & Science: Digital Histories,” an initiative of the Max Planck Research Group ”Epistemes of Modern Acoustics.” (https://acoustics.mpiwg- berlin.mpg.de) The ”Sound & Science” database, currently in its beta version, provides valuable access to a diverse range of sources in the history of acoustics-many of them previously unavailable in digital format. The resource includes a multimedia archive of primary source material, documentation of surviving technology, and historical reenactments of crucial experiments in acoustics. The database can be browsed using a full-text search function or via curated collections, or resources can be viewed on a map. In my paper, I will outline user-scenarios for the database’s visualization tools, which open up large-scale and longue durée views on the history of sound and science, allowing users to explore and assemble material in unprecedented ways.
	The Consolidation of Engineering Acoustics as an Example of Contextual History of Science Joachim Scheuren [Abstract] Following widespread practice in natural sciences, the history of Acoustics is often understood and depicted as a sequence of scientific ideas and achievements provided by ingenious individual scientists or scientific schools. Even if this approach is accepted because it foregrounds the sequence and relation of ideas it risks to ignore the driving energy of any social and technical context of science. This is particularly true for applied sciences like acoustics which, from a certain level of scientific insight, are much more driven by the needs of technology than by purely scientific interest and progress. The mutual influence and dependence of scientific, technological and social determinants can be illustrated exemplarily for the renaissance of acoustics in the mid of the last century. Being underestimated as conclusively understood and lacking new challenges, the relevance of Acoustics changed significantly with the demands of the post-war era. The respective requirements were able to initiate an iterative process linking technological progress and needs with new scientific insight. With reference to some typical examples it will be shown how these interactive efforts finally established the disciplines of engineering acoustics and noise control which essentially contributed to renew and strengthen acoustics as an indispensable interdisciplinary field.
	Echoes and Guns: Objects, Practices and Aims in Early Modern Acoustics Leendert Van Der Miesen [Abstract] As interest in acoustical experiments grew in the early modern period, the range of acoustic objects and practices greatly expanded. Although many acoustic objects-such as the string or organ pipe-were still related to the field of music, scholars increasingly studied the acoustic properties of other objects, for example guns or the echo. These objects played a large part in the development of notions of sound propagation and indicate a wide variety of uses and aims of acoustical knowledge. Through examples ranging from the French scholar Marin Mersenne in the early seventeenth century to scientific academies such as the Royal Society in London and the Académie des sciences in Paris, my paper considers the different sites of experimentation, and especially the aims and overarching practices of the nascent discipline of acoustics. With these objects, I am able to investigate the different roles of experimentation and the possible implementation of acoustical knowledge.
	The History of Acoustics in Breslau/Wroclaw before and after 1945: A Bridge over Time Andrzej Dobrucki, Peter Koeltzsch [Abstract] The paper presents the history of research and the most important people active in the field of acoustics in Breslau before 1945 and in Wrocław after this date. The first important acoustician to lecture in Breslau and here died in 1827 was Ernst Chladni. Organized research in acoustics originated at the end of the 19th century by Otto Lummer at the University of Breslau. His successor was Professor Clemens Schaefer. In 1910, the TH Breslau was founded, and research in acoustics was also conducted here. The leading figure was Erich Waetzmann. He dealt with room acoustics and electroacoustics. Professor Kurt Schuster was a pioneer in the measurements of acoustic impedance. The book ”Ultrasound” by Prof. Ludwig Bergmann remains important to this day. After 1945, the research was concentrated at the Wrocław University of Technology. The pioneer of this research was Professor Zbigniew Żyszkowski, who dealt with electroacoustics and telephoneometry. His students were: Wojciech Majewski, specialist in speech and Janusz Renowski - psychoacoustician. The professors in next generation are Andrzej Dobrucki, who currently heads the Chair of Acoustics and Tadeusz Gudra, active in ultrasound. The youngest professor, from the further generation is Krzysztof Opieliński.
	The Church building secularization through its sound Sound in a contemporary case study Maria Cairoli, Livio Mazzarella [Abstract] Acoustics in a contemporary liturgical place, particularly in a catholic context, is far from obvious. This work describes some main indications for the design of new churches, starting from the project of Sant’Ambrogio Church in Trezzano, Milan (IT). The article presents room acoustics procedures, results, discussions, theoretical interpretations of the sound propagation in the specified space, and the implementation of computer simulation techniques, emphasizing the continuity and breaking elements with the historical churches. In the past the sacred sense in a church was linked to emphatic rites and to the attainment of transcendent states. This effect had to be greatly reinforced by the surrounding atmosphere created in very reverberant spaces and possible echoes that infatuated the perception of God and his presence. On the contrary, nowadays concentration and intimacy are preferred, the reverberation time is reduced, increasing the speech intelligibility of the homily to understand the message of God, while music or songs steel accompany the choral word of church goers. The proposed acoustic design includes a new approach with an acoustic response suitable for both speech and music, considering also the recent room acoustics studies in other typologies of building such as congress centers and multipurpose auditoria.

Organization:

B. Boren, American University A. Andreopoulou, Laboratory of Music Acoustics and Technology, University of Athens

	Acoustic effects at prehistoric landscapes: an archaeoacoustics analysis of rock art sites from Western Mediterranean Margarita Díaz-Andreu, Angelo Farina, Enrico Armelloni, Laura Coltofean, Mathieu Picas, Tommaso Mattioli [Abstract] Many pre-state societies around the world give special importance to places where special acoustic effects are generated, and often these places receive special treatment including the production of rock paintings in them. In the Western Mediterranean, it seems that outstanding acoustic effects such as directional echoes, augmented audibility and long reverberation time are present in some rock art areas with Neolithic depictions dating between the VII and IV millennia BC (painted in Macroschematic, Levantine and Schematic rock art styles). On the basis of this evidence, it has been argued that there is a strong probability of acoustics having been used as a method by Neolithic artists to select the shelters in which to produce rock art. This paper presents the results of the ongoing ERC research project in archaeoacoustics (Artsoundscapes - The sound of special places: exploring rock art soundscapes and the sacred). This project seeks to explore the role of sound in the creation and use of these archaeological sites. In particular, the authors discuss the results of previous fieldwork in three countries (Spain, France and Italy) and the development of an innovative set of research methods that includes 3D Ambisonic recordings, Transmission Loss measurements, and GIS soundshed analysis.
	From Methodology to Archaeoacoustics in the Time of Scripture: Complex Dialogue Between Archaeological Evidence, Texts from Scholars and Written Mentions Jean-Christophe Valiere, Estèle Dupuy, Bénédicte Bertholon [Abstract] Nearly 15 years of research on the acoustics of pots from 11th to 18th centuries have led us to crossing data from various disciplines and developing a specific methodology. The presentation analyzes the pitfalls of interpreting this technique through the eyes of the scholars that have addressed the issue since antiquity and of written mentions rather than using the archaeoacoustical data found in France and Europe. An archaeoacoustic research will only be meaningful if it can prove a deliberate intention to produce an object with an acoustic function. However, such a proof necessitates observing regularities, by studying several buildings or objects that attest that an effective choice has been made. A common error is to restrict the hypotheses to those which are scientifically admissible today. The case of pots acoustics is particularly edifying in this respect. Nevertheless, precise analysis of historical texts makes it possible to widen the field of hypotheses. Semantic study of the evolution of the meaning of key words and their associated concepts makes it possible to study the evolution of the acoustical techniques and to observe their progress through the centuries.
	A case of archaeological evidence in favour of acoustical intentions linked with pots in church vaults: Montivilliers Abbey Jean-Dominique Polack, Jean-Christophe Valiere, Bénédicte Bertholon, Pauline Carvalho [Abstract] Montivilliers Abbey near Le Havre in France presents a case where pots in the vault were obviously employed to act on the acoustics. The archives of the abbey contain a written mention that links pots with acoustics. Originally built in the 13th century, the Romanesque vault of the transept crossing was later modified at an unknown date by the addition of pots. In 1648 the vault was lowered down, hiding the original pots. A dozen of new pots were inserted in the new vault. The presentation focuses on the acoustics of the abbey and documents the organization of the pots by the builders in the 1st and the 2nd vaults. Room-acoustical measurements of the church were also carried out in order to evaluate its acoustical quality. A major conclusion is that lowering the vault has singled out the church choir where the nuns chanted and which is characterized by reverberation times lower than in the nave. Intermediate reverberation times were measured in the crossing. Reverberation times in octave 125Hz, which includes the resonance frequencies of the pots, are marginally lower in the choir and the transept for source in the nave, which could indicate an effect of the pots.
	Virtual reconstructions of the Théâtre de l’Athénée for archeoacoustic study David Poirier-Quinot, Barteld N. J. Postma, Brian F. G. Katz [Abstract] The French ECHO project is a multi-disciplinary collaboration combining the efforts of historians, theater scientists, and acousticians to study the use of voice in 20th century French theaters. An audio-visual simulation framework was created which combines auralizations with visualizations of former configurations of the Théâtre de l’Athénée, Paris. These configurations corresponded to a series of renovations, enabling researchers to perceive realistic stage performances in foregone rooms. Simulations include the room, two actors moving on stage, and an audience. To achieve these simulations, archived architectural drawings were studied, providing various details of the different theater configurations from which the corresponding visual and room acoustic geometrical acoustics (GA) models were created. The GA model of the theater’s current state was calibrated to reference measurements. Subsequent modifications to the GA model allowed for reversing several renovations to arrive at the original 1896 configuration of the Athénée. Simulations across configurations accommodated the investigation of the impact of these renovations on acoustic conditions, such as reverberation and clarity. These observations can then be compared to historical information, such as theater critics of the time. The final resulting audio-visual simulations allow for 360∘ presentations at various positions in the theatre using commercial standard hardware.
	Acoustic Simulation of Elizabeth I's Speech at Tilbury Braxton Boren [Abstract] The largest gatherings in human history prior to the advent of electronic amplification were all effectively limited by the loudness and intelligibility of the human voice. Various accounts throughout history depict single orators addressing crowds or armies of many tens of thousands, though some historians have cast doubts onto the veracity of these claims. Prior acoustic simulation has shown that reasonable intelligibility might be achieved for up to 20,000 listeners for very loud orators and very quiet crowds with high density. Nearly all such accounts, however, depict a man addressing the crowd, with one notable exception: in 1588 Elizabeth I of England addressed a large army of between 8,000-19,000 troops gathered to repel the impending invasion posed by the Spanish Armada. The acoustic plausibility of this account is tested via STI simulations, focusing primarily on the density, topography, and environmental conditions at the time of the speech, as well as the frequency-dependent differences between female and male orators’ vocal spectra and a sound radiation patterns.

Organization:

M. Gatt, Technical University of Munich

	How does what we hear sound? The qualia problem in acoustics Michael Haverkamp [Abstract] Development of quantitative methods for description of auditory perception has extended the possibilities of detailed analysis and optimization of sound properties. Psychoacoustic quantities help to assess audible features like loudness, roughness, sharpness, tonality and pitch. But why is it still not possible to precisely describe how an auditory phenomenon really manifests itself within the subjective perception of an individual? This would be of essential interest in terms of definition of targets for sound design, legal determination of branding sounds and comprehension of annoyance of auditory incidents. In order to approach this question, the so called qualia- problem shall be outlined in view of auditory perception. It is evident that development of effective metrics for sound evaluation is hindered by this matter of fact. Even under the hypothetic presumption that a complete set of relevant parameters could be achieved, a definite description would fail due to the impossibility to include the perceived subjective content. Moreover, the holism of perception impedes to achieve this goal. Various approaches will be discussed briefly: 1. Verbalization of sensations and onomatopoeia 2. Inclusion of further modalities as tertium comparationis 3. Parameters related to the human body, like movement 4. Approaches for notation of music and sounds
	Alexa, Siri and more: The impact of speech recognition on social behaviour and our responsibility as its creators Andrea Ebner [Abstract] Digitalization and globalization allow us to create machines that can be used all over the world. But we must consider what people in the target markets want. What do they think of the hidden microphones in their living rooms, why are they afraid of speech-control? Are people ready to give us their trust in our devices and their consent to share their privacy with us? Since the catchphrase Artificial Intelligence can be read in every newspaper and it seems like there is no other technical topic out there to talk about any more, I want to ask some questions. What are the society’s worries? Where may they come from? How can we handle our responsibility as the creators of change? The political debate often seems idealistic. We as scientists and engineers can support the decision-makers and the society with giving them rational facts and provide simple, understandable explanations of what is going on in the tech world. Because it is not just about with which good or evil device we interact. Instead of being afraid of change it is a matter of which principles we want to manifest this change on to transform the world.
	The inner voice Monika Gatt, Marcus Maeder, Steffen Marburg [Abstract] With inner voice, contemporary common sense often labels an individual creative motivation of acting in private life, social media and business. From a philosophical point of view, the inner voice refers to the ancient Socratic conception of soul and genius. And that is why the secret roots of the profession ”Ingenieur” (german) and Engineer (english) and the understanding of audition might be found deaper: deaper in culture, in science and as well in someone’s becoming a person. On the one hand, reliable results of a research study in Acoustic focus on criteria of objectivity, but every step might also be mesuared by each one’s own personal situation. This talk deliveres three main topics: 1. a crucial philosophical-logical matrix of objectivity as an example of subjective resonance in research, 2. a phenomenological diagnosis of privacy space in relation to outdoor noise, 3. a free philosophical development of the inner voice as a healthy ethical space of narration.

Organization:

S. Luzzi, Vie en.ro.se. Ingegneria C. Bartalucci, University of Florence A. Di Bella, University of Padova -Department of Industrial Engineering

	Comparison of the recreational noise regulation in Russia, Europe, and Asia Sergio Luzzi, Chiara Bartalucci, Iuliia Rassoshenko, Aleksandr Shashurin, Nickolay Ivanov [Abstract] In October 2018 the new Environmental Noise Guidelines for the European Region were published by the World Health Organization (WHO), providing strong evidence that noise is one of the top environmental hazards to both physical and mental health and well being in Europe. However, recreational noise has not yet been investigated in detail. After a preface about the more recent outcomes on recreational noise provided by EU Guidelines and projects, the current article presents the requirements for noise levels in recreational areas of various categories (parks, specially protected areas, reserves, etc.) in Russia, Europe and Asia. Moreover, it describes the main existing legal and regulatory acts in the countries, the history of their development and restrictions on their use. Furthermore, the article shows a comparative analysis with the allocation of advantages and disadvantages of each system. Finally, suggestions have been made to improve the system for regulating noise levels in recreational areas.
	Recreational Noise Management in Turkey Nılgün Akbulut Coban, Sezer Kaya, Güray Doğan, Mustafa Çoban [Abstract] Noise is one of the inevitable environmental types of pollution especially in densely populated urban areas with the design approach including the recreational places within the residential areas. It is well-known that there are lots of evidence on the health effects of environmental noise and to control this, the improvements have been carrying on in both legislative basis and scientific basis. In Turkey, during the harmonization of European Directives into national legislation as a full candidate country of European Union, 2002/49/EC of Environmental Noise Directive was one of the first ones transposed with the name of ”Regulation on Assessment and Management of Environmental Noise (RAMEN)” in 2005. In the scope of RAMEN, the noise criteria and evaluation procedures are identified for the main noise sources including also entertainment places. In this paper, recreational noise management in Turkey will be explained and discussed through giving the information on what the main environmental noise criteria for entertainment places is, which authorities responsible for noise audits are, how the procedures to take the music allowance are applied, what requirements of technical staff to conduct the environmental noise measurements and preparing the noise assessment reports are needed.
	Long term monitoring of noise pollution in social gatherings places: time analysis and acoustic capacity as supports of management strategies Enrico Gallo, Louena Shtrepi [Abstract] City of Torino (Italy), as many European cities, is facing an increasing noise pollution related to recreational noise of nightlife in open urban areas, such as streets, squares and terraces, where thousands of people meet spending all evening and night time. Noise from people speaking in these spaces can be very loud, due to crowd levels (since communication is possible only with a raised voice level) and people behaviour. This has led administrations to attempt to manage this specific noise problems with different strategies. This study shows the analysis of long term noise data collected in two years by a low-cost IoT network in Torino. It highlighted the regularities on night levels, mainly on a weekly basis and with seasonal trends. Preliminary investigations on time series have been developed, in order to correlate noise levels and number of people involved, coupling environmental data and crowd sensors. Moreover, the concept of ëAcoustic Capacity’ related to the noise levels has been investigated. This approach based on a deeper quantitative knowledge is explored as an help to local administration and stakeholders in planning and implementing mitigation actions.
	Revision of Chinese National Standard GB 50118 - Code for Design of Sound Insulation of Civil Buildings Guojun Yan, Weibin Wu, Jie Lin, Chun Xu, Qiyuan Zhao [Abstract] The national standard GB 50118 Code for Design of Sound Insulation of Civil Buildings, issued by Ministry of Housing and Urban-Rural Development (MOHURD), is the basic code for indoor acoustical quality of civil buildings in China. It stipulates the limits for indoor noise levels and sound insulation performance of building elements, and specifies the design criteria of noise control. The current version was published in 2010, in which requirements had been raised compared to previous version. However, with the rapid development in China during the last decade, current regulations could not meet the growing needs for better sound environment. Traffic noise, HVAC noise and structure-borne noise from service equipment have become increasingly annoying, and inadequate sound insulation aggravates the situation. Therefore, MOHURD commissioned China Academy of Building Research to undertake the revision of current standard in 2018. Based on the correlation between field measurement results and occupants’ perception, and referring to guidelines and regulations in developed regions, we’d like to put forward stricter requirements for indoor noise and sound insulation performance of building elements. Besides, limits for structure-borne noise caused by vibrations of building service equipment or any other noise sources will be introduced to assess the dominant low-frequency noise.

Organization:

G. Brambilla, IDASC-CNR G. Licitra, ARPAT – Environmental Protection Agency of Region of Tuscany, Area Vasta Costa C. Asensio, I2A2-Universidad Politécnica de Madrid B. Schulte-Fortkamp, Technische Universität Berlin

	Acoustic noise maps - exchange / integration of data within various database systems - problems, needs, new possibilities, interopracy Tomasz Malec [Abstract] When creating acoustic maps, it is necessary to use many data sources such as Numeric Model of Terrain (NMT), Building Data Base (BDT), Environmental Data Base (cover, bridge, etc.) and others. For these data it is necessary to apply a layer of acoustic data such as traffic parameters and type of pavement for roads/rails, sound power level for industrial source, noise barriers, etc. The sources of these data may vary, including those based on open licenses(OpenStreetMap, Inspire), paid government data (NMT, BDT, etc.). Each of the above data can be presented in several generally used data formats. All data was integrating in the noise map calculation programs. This approach blurs the transparency of data. It becomes legitimate to ask if it can be standardized so that it is possible to pass data between programs to calculate noise maps or generally available GIS tools. The work presents an overview of data, tools for their usage, methods of their storage. Based on the review, the directions of changes necessary to increase the transparency of these data and their interoperability were proposed.
	Noise Low Emission Zone implementation in urban planning: results of monitoring activities in pilot area of LIFE MONZA project Rosalba Silvaggio, Salvatore Curcuruto, Manlio Maggi, Antonio Fasanella, Giorgio Cattani, Alessandro Di Menno Di Bucchianico, Alessandra Gaeta, Gianluca Leone, Enrico Mazzocchi, Raffaella Bellomini, Sergio Luzzi, Francesco Borchi, Chiara Bartalucci, Monica Carfagni, Giulio Arcangeli, Nicola Mucci, Carlo Casati, Giulia Pessina [Abstract] The main objective of LIFE MONZA (Methodologies fOr Noise low emission Zones introduction And management), started in September 2016, concerns an easy-replicable method, and related guidelines for regulatory purposes, for the identification and the management of the Noise Low Emission Zones, urban areas subject to traffic restrictions, whose impacts and benefits regarding noise issues are testing in the pilot area of the city of Monza, located in North Italy. Noise LEZ has been introduced in Libertà district, by top-down actions (traffic restriction and vehicles speed regulations, lanes-width reduction and pedestrian crossing introduction, substitution of the asphalt with a silent road surface) and encouraging an active involvement of the people, in the definition of a more sustainable lifestyle (bottom-up actions). Monitoring activities in pilot area, regarding noise, air quality improvement and the wellbeing conditions of the citizens, have been carried out, and they are still ongoing, before and after the Noise LEZ introduction and the available results are described in this paper, in order to discuss and to evaluate, in an unitary way, the main effects of Noise LEZ introduction in urban planning and in lifestyle of inhabitants.
	Soundscape Planning as a Tool for Urban Planning Moritz Lippold, Bryce Timothy Lawrence [Abstract] Dealing with noise is one of the major challenges cities are facing. Spatial planning and noise legislation in Germany is commonly based on the idea of noise avoidance and noise control, whereby noise is treated as a waste product. At the same time, there is consensus that quantitative noise reduction is not enough to address the acoustic environment. Therefore a readjustment of the instruments and methods of spatial planning is necessary. As a suitable alternative in thinking about noise the soundscape approach might open up new, more holistic perspectives by considering the acoustic environment as a resource. However, there is currently a lack of practical approaches to include the idea of Soundscapes in spatial planning. This study asks the question: How can a soundscape planning approach be integrated as an instrument for spatial planning in Germany? To answer the research question, a systematic review of existing literature and a summary of German noise-related policy instruments will provide a framework of how sound and noise is treated in Germanys legal system and give insight into how a soundscape approach could be integrated into the German spatial planning system.
	Noise complaints and its relation to socio-economic factors at city/region scale in England Huan Tong, Jian Kang [Abstract] Reporting noise complaints is an important part of informing noise legislation. While the effects of various socio-economic factors on noise perception have been investigated in previous studies, the aim of this study is to examine relationships between the rate of noise complaints and socio- economic factors. The noise complaints and socio-economic datasets from the government open data source at district and unitary authority levels across the England are used. The socio-economic dataset is categorised into four groups in this study, namely demographic, working, property and deprivation factors. Correlation analysis is conducted between noise complaint rate and socio-economic factors, and the results suggest that the correlations are generally significant. Cities/regions with a higher proportion of young and single residents are likely to receive more noise complaints, and so are city/regions with a higher unemployment rate and higher proportion of residents living in flats. The deprived city/regions, measured by the English Index of Multiple Deprivation, tend to have higher noise complaint rate.
	Introduction to a new approach in urban acoustics Alexander Lee, Jochen Schaal, Berndt Zeitler [Abstract] Statutory rules describe how to generate noise maps by defining methods and the choice of input data for calculating sound propagation. In highly reflective and diffractive environments, as exist in urban surroundings, those rule based methods have their limits. The currently used geographical input data for buildings consists of generalized blocks, which disregard detailed structures and materials in the cities’ streetscape. However, for urban environments a new simulation approach, applied to an urban model with more structural details, could be useful for planning and predicting noise. Luckily, building data with a higher level of detail (LOD) is available through geographic information systems, though cannot be utilized by conventional noise map calculations. So, as a first step, a conventional noise map of such a low LOD input is compared to a map created by soundPLANnoise’s sound particle diffraction (SPD) model with the same data set. Since the SPD model can incorporate high LOD input, in a second step, the effect of model details on noise maps is compared. Stöckach will serve as showcase area, a quarter in Eurocities member city, Stuttgart.

Organization:

F. Schelle, Institute for Occupational Safety and Health (IFA) Y. Demiral, DOKUZ EYLUL UNIVERSITY

	ASR A3.7 “Noise”: Technical Rules for the German Ordinance on Workplaces Florian Schelle, Jan Selzer [Abstract] The most fundamental national requirements regarding noise at workplaces in Germany are the ”Ordinance on Noise and Vibration Protection” for workplaces characterized by an LEX,8h ≥ 80 dB, as well as the ”Ordinance on Workplaces”, which applies to workplaces with LEX,8h < 80 dB. Both are concretised by means of ”Technical Rules”. Important aspects of both Technical Rules will be presented. An emphasis will be put on the relatively new ASR A3.7 ”Noise”, which came into force in May 2018. It was designed especially for the assessment and rating of noise that is not hazardous to the inner ear, but can potentially cause extra-aural effects. Therefor it relies on reverberation time along with a so-called ”Rating Level” Lr, which will be presented in more detail. The given parameters, limit values and recommendations will be discussed as well as practical implications based on the experience of actual workplace measurements.
	Noise emission data as a prerequisite for Buy Quiet: Challenges related to the European Machinery and Outdoor Noise Directive Fabian Heisterkamp [Abstract] Article 5 of the OSH Directive 2003/10/EC on noise aims at ”avoiding or reducing the exposure” of workers to noise, so that (Article 5 1.) ”...the risks arising from exposure to noise shall be eliminated at source or reduced to a minimum”. The Buy Quiet-concept, i.e. the selection of machinery with a special focus on low noise emissions, can represent an effective means to comply with this legal requirement. However, reliable and useful noise emission data are not always available. Especially, machinery designed to be used outdoors can represent a challenge. This kind of machinery is covered by both the ”Outdoor Noise”-Directive 2000/14/EC (OND), a Global Approach directive, and the Machinery Directive 2006/42/EC (MD), a New Approach directive. The interplay of these two directives results in noise emission declarations that differ from those of other machinery and can represent a challenge for manufacturers as well as employers. The arising problems are discussed and explained at observations in practice. Equipment covered by the OND is almost always provided with emission data. The challenges when trying to select quieter machines based on these data are discussed and possible strategies to identify quieter machines, but also to improve the OND, are presented.
	Admissible Values and Methods of Measurement of Noise, Ultrasonic Noise and Infrasonic Noise at Workplaces in Poland Dariusz Pleban [Abstract] The adverse effects of noise, ultrasonic noise and infrasonic noise on human are associated mainly with the hearing organ. Nevertheless, non-auditory effects can be also observed in the entire human body. Therefore, the admissible values of noise and ultrasonic noise at workstations are specified in the Regulation of the Minister of Family, Labour and Social Policy. According to this Regulation, measurement methods should be specified in Polish Standards, and, consequently, the method for the measurement of noise at workplaces is defined in the Polish Standard PN-EN ISO 9612. In the case of ultrasonic noise the draft Polish Standard prPN-Z-01399 ”Ultrasonic noise. Requirements for measurements in the work environment” was developed on the basis of the results of studies carried out in CIOP-PIB. Moreover, as regards the workplaces occupied by juveniles and pregnant women, the admissible values of noise, ultrasonic noise and infrasonic noise are defined in regulations issued by the Council of Ministers. In turn, values that constitute the criteria of annoyance of noise and infrasonic noise are defined in Polish Standards PN-N-01307 and PN-N-01338. The discussion of the admissible values and the characterization of the draft Polish Standard prPN-Z-01339 are presented in the paper.
	Ultrasound Noise Policy and Assessment: Canada Safety Code – 24, a Canadian Perspective Lixue Wu [Abstract] Ultrasound is extensively used nowadays in work and home. Scientific evidence on the health effects of exposure to ultrasound recommends that caution should be taken when using ultrasound. This leads to safety regulations established and enforced by government agencies in many countries. In Canada, early regulations respecting occupational safety and health made under Part IV of the Canada Labour Code, were substituted by Canada Occupational Health and Safety Regulations (SOR/86-304), effective March 31, 1986. Where a device that is capable of producing and emitting energy in the form of non-ionizing radiation is used in the work place, the employer shall implement the applicable document, in respect of ultrasound equipment, Safety Code - 24, dated 1990. The National Research Council Canada (NRC) is one of the government agencies who provided reviews and comments in the preparation of the Code. In this paper, the Canadian recommended human exposure limits for airborne ultrasound and its measurement techniques are discussed with the focus on challenges. To implement the safety code, a measurement system for ultrasound pressure level measurements was developed recently at NRC. This paper presents the details with the focus on measurement uncertainties. An onsite noise assessment example is given using the system.
	Noise exposure of employees in retail trade Jan Selzer, Florian Schelle, Andrea Wolff, Frank Rokosch, Anne Gehrke [Abstract] Noise can lead to extra-aural effects, which can diminish the health of employees in the long term. To protect the employees from extra-aural noise effects, a new national regulation for measurement and rating of noise at workplaces became active in Germany by means of a technical rule. In order to assess the situation of employees in retail trade, measurements of noise exposure were carried out following the mentioned technical rule and ISO 9612. Personal sound exposure meters were used to determine the physical exposure at the ear of employees. Stationary measurements next to the cashier’s area were performed additionally. For psychological evaluation of working-condition-perception, a validated questionnaire (BASA II) was used. Furthermore, room acoustic measurements were performed in representative retail stores as well as binaural recordings for psychoacoustic analysis purposes. Outcomes of this project will be presented and discussed for personal and stationary sound pressure levels. Measurement method and rating of results in German legal requirements turned out to be difficult to apply for these types of workplaces, as will be discussed in detail. In addition, the results of the measurements and the questionnaire will be reviewed to show the relationship between perceptual and analytical variables.

Organization:

H. Boegli, Federal Office for the Environment R. Weinandy, Umweltbundesamt

	The EPA-Network Interest Group of Noise Abatement (IGNA) Urs Walker, René Weinandy, Hans Boegli, Nina Mahler [Abstract] Noise abatement in Europe has made considerable progress in recent decades. Starting with the Green Paper on Noise Policy in 1996, followed by the adoption of the Environmental Noise Directive in 2002, many noise maps and action plans to reduce noise exposure of the population have been developed. However, the problem of noise has not yet been solved and that is why the EPA Network - the informal group of ”Heads of European environment protection agencies” - decided in 2010 to boost this topic by creating an ”Interest Group on Noise Abatement” (IGNA). IGNA’s first mandate from 2011 to 2016 focused on best abatement measures at traffic noise, decision criteria for applying noise abatement measures, limit values and monitoring methods. IGNA organised eight workshops and issued four reports covering road, railway and aircraft noise as well as various letters to EU-institutions asking for tighter noise regulations. In its second mandate from 2017 to 2022 IGNA concentrates on all noise sources, limit values in Europe and quiet areas. The activities of IGNA are leading to a common understanding of how to tackle and solve noise problems and to disseminate the information to players and stakeholders in other countries.
	Recommendations for Traffic Noise Abatement René Weinandy, Hans Boegli, Urs Walker [Abstract] Since 2010 the EPA-Network Interest Group on Noise Abatement has developed various reports with best practice recommendations covering road, railway and aircraft noise as well as decision criteria for applying noise abatement measures. The most important recommendations are as follows: Road traffic: Further tighten limit values for road vehicles and tyres, extent the knowledge of the public (both users, suppliers and other stake holders) about the noisiness class of products like vehicles and tyres, continue development and application of low noise surfaces. Railway traffic: substitute cast-iron block brakes with even synthetic material or disk brake and maintain high quality wheel and rail smoothness, introduce stricter noise requirements for all existing freight wagons. Aircraft traffic: Introduce tighter limit values of noise certification levels, harmonize noise categories of aircraft, establish permanent incentives to optimize flight procedures regarding noise. Criteria for applying abatement measures: Develop harmonized methods to determine and internalize the costs and benefits of measures to reduce traffic noise. Despite the best practice list, one has to be aware that the classic noise abatement procedure has to be complemented with new innovative measures in order to face the challenge of future trends and developments. IGNA is working on that.
	Problems of Road Traffic Noise Annoyance and Sleep Disturbance in Some Slovakian Cities Ladislav Mihalcik, Stanislav Sekretar, Jana Jurkovicova, Jan Simonovic, Lubica Argalasova [Abstract] The paper presents the results of a pilot cross-sectional study focused on subjective traffic noise annoyance and sleep disturbance among the residents living close to major inner city corridors in several Slovakian towns including the capital Bratislava. The subjective adaptation to traffic noise, subjective evaluation of health status were assessed as well. The questionnaire survey was distributed to the pilot sample of 543 respondents (average age 45,3 (73,1% respondents from 35 to 65 years), 57% females, 71% living in houses more than six years). The inhabitants with bedroom windows facing noisy streets or quite streets represented the exposed and the control group. Road traffic noise annoys significantly more daily and night activities of respondents in the exposed group, who are unable to adapt to it (OR=5,28; 95 % CI=3,56−7,86 for sleep disturbance). The questionnaire was supplemented by traffic noise- measurements in Bratislava on noisy facades (LAeq,day = from 57 dB to 72 dB, LAeq,night= from 52 dB to 62 dB) close to major streets with intensity transport from about 16000 to 38000 vehicles per day. After completion of the results, we plan to propose interim measures to new apartment buildings as well as intervention procedures.
	Overview of environmental noise limits in the European Region Rosan Nusselder, Bert Peeters [Abstract] In October 2018, the World Health Organization (WHO) published environmental noise guidelines based on dose-response curves for relevant noise-related health impacts. The guidelines provide recommended noise limits, but do not describe what consequences of exceedance should be. Also, it is helpful to know how the currently adopted limit values in European countries relate to the recommended limit values. Our research aims to address these questions. To do so, a survey has been sent out to national experts to collect data regarding current critical noise limits. From the output, an overview has been created which allows a comparison of limit values between countries. Also, a comparison is made with respect to the new WHO recommendations, revealing the current state of noise legislation in the European region. In addition, information has been gathered on the scope and basis of these values, as well as detailed information on their assessment, exceptions and legal consequences. From it, general trends in noise level policy and enforcement methods are extracted. The results provide insight in the broad spectrum of noise legislation within the European region, and form a basis from which to build best practices for countries that aim to implement the WHO guidelines.
	Assessment of Quiet Areas in Europe Eulalia Peris, Nuria Blanes Guardia, Jaume Fons-Esteve, Miquel Sáinz De La Maza [Abstract] Living in a quiet area provides health benefits such as increased quality of life and well-being. Nearby access to quiet areas can also offer psychological restoration and can help in reducing noise annoyance reactions. Quiet areas are not only beneficial for human health but also help to protect areas of valuable habitat. The Environmental Noise Directive (END) recognises the need to preserve areas of good acoustic quality, referred as ”quiet areas”. However, the data reported as part of the END contains little information on how the countries, regions and cities define and protect quiet areas in their territories and whether there has been a significant improvement in designation and protection of these areas over the past years. The aim of this paper is to present an overview of how countries define and protect quiet areas in Europe. This is achieved using data from case studies comprised of online questionnaires, which were completed by noise representatives of different countries, regions and cities. In addition to the analysis of the questionnaires, a combined spatial assessment of noise exposure, land use and land cover data for areas potentially unaffected by noise pollution in European cities is presented for 2012 and 2017.
	Noise Advocacy Tony Paul Dolan [Abstract] Following the completion of the Round 3 noise maps and action plans, the Irish Environmental Protection Agency is focussed on the provision of public information and advocacy around transport noise. The publication of the noise maps was tweeted by our Communications Unit. The noise maps are being used by the various Local Authorities to communicate to the public about the modelled noise levels in their area and how they may be addressed in their noise action plan. It has given the public improved access and a better understanding of what the noise maps are all about. We are also developing a noise briefing paper to give an update on environmental noise, which we will add to the noise mapping information on the EPA website. In addition, we are producing an environmental noise Factsheet/Infographic to help provide better public information on noise. We are also liaising with our Communication team to publish a noise feature article, and some other noise related features during 2019. The intention is to increase public awareness around environmental noise with a view to putting a greater focus on the various LA noise maps and action plans, and their implementation over the next five years.
	New trends in noise abatement Hans Boegli, Urs Walker, René Weinandy [Abstract] The classic noise abatement approach consisted in evaluating the problem by establishing noise maps, evaluating noise abatement measures in form of action plans and keeping the public informed of the progress with a regular monitoring. This was at least the procedure undertaken in Switzerland over the last forty years and the EU follows basically the same procedure. This approach is driven by the belief that there are some efficient technical or constructive measures available to reduce noise emissions. Moreover, one expected that low-noise innovations would inevitably solve the problem and noise as an environmental burden would eventually disappear from the agenda. However, new trends and developments have made noise abatement aiming at moving targets and facing new battlefields of conflicts. These changes can be attributed to evolution of attitude of human beings, social-demographic changes and progresses in the techno-economical environment. In answer to these developments, the strategy of noise abatement has to be adapted in order to cope with the new challenges. This adaptation should ensure that type of noise abatement measures are less ”end-of-pipe” such as noise barriers or protection windows, but measures at source based on holistic concepts that integrate all relevant aspects of modern life.

Organization:

B. Peeters, M+P A. Kok, RIVM

	Flaws in the Cnossos Calculation Method and Proposed Solutions Arnaud Kok [Abstract] January 1 2019 all member states were required to have transposed the Cnossos calculation method (EU directive 2015/996) in their legislation. In preparation a study was conducted in 2017 to evaluate the new calculation method. Part if this evaluation was to see if the method could be used as a replacement for the Dutch national method. In this study we found errors within the method that, if implemented, would lead to implausible results. The results of this study were presented to the EU commission and the Noise Regulatory Committee. As a result an EU working group, chaired by the Netherlands, was established to study and propose amendments to the method. This EU working group has found numerous issues. Some issues are about unclear text, that could lead to different interpretations of the method. Other issues are more fundamental. They are clear errors in the method. One example is a problem that occurs with multiple diffractions in favourable conditions. For almost all the issues a solution was drafted and a report was finalized where these issues and proposed solutions are presented. In this paper we will address the problems with the method and proposed solutions.
	Establishing noise emission for an electric road vehicle category in CNOSSOS-EU Bert Peeters, Erik De Graaff, Hilary Notley, Simon Shilton, Matthew Muirhead [Abstract] The CNOSSOS-EU road noise model described in EU Directive 2015/996 makes use of model coefficients that represent the average noise emission spectrum and speed dependence for each particular vehicle category. Separate coefficients are provided for the road surface influence and for accelerating/decelerating at intersections. Appendix F provides default values for cars, trucks and two- wheelers with combustion engines. This study describes a method to establish noise emission coefficients and road surface corrections for a new vehicle category. The primary aim is to include (hybrid) electric vehicles in the noise assessments, to enable evaluation of EV/HV proliferation scenarios. To separate rolling and propulsion noise, a combination of pass-by (CPB) and close proximity (CPX) methods on an appropriate selection of instrumented vehicles is proposed. Acceleration and deceleration coefficients may also be derived from instrumented vehicle measurements. The total noise emission is calibrated using roadside measurements on vehicles in regular traffic. Acoustic transfer functions are calculated to derive the sound power coefficients from the measurement results. The methodology provides an approach to adding any vehicle categories to CNOSSOS-EU, and for a potential future BSI standard to replace the existing British CRTN assessment method. Recommendations for further development of such a standard are given.
	The determination of road surface corrections for CNOSSOS-EU model for the emission of road traffic noise Fabienne Anfosso Ledee, Luc Goubert [Abstract] In the CNOSSOS-EU noise emission model for road traffic noise, the calculation of sound power emission of vehicles introduces a road surface correction term, which can be calculated from a set of values provided for non-standard types of road surfaces. However, the data may be of limited use for the authorities in charge of strategic noise mapping if they do not match with the national or local types of surfaces or if newly developed types of surfaces are laid on their networks. A procedure for the acoustic characterization of road surface properties is currently discussed within the European standardization group CEN/TC227/WG5. This procedure, based on CPX tyre-road noise measurements, could serve in two ways for the evaluation of road surface corrections as input parameters in CNOSSOS-EU. The first is a direct snapshot measurement on an existing network, providing a rapid and up to date evaluation of the road surface correction term, without considering the nature of the surface. The second is a reference measurement to collect reference data to serve for calculations at later stage. The paper will present the proposed method and its compatibility with the existing noise emission model.
	Transposition of CNOSSOS-EU into German Law René Weinandy, Thomas Myck [Abstract] One of the most important environmental issues in densely populated areas is the problem of noise. Traffic noise from cars, railway vehicles and airports located in close proximity to the city is not only annoying for residents; it also leads to serious health issues and has an enormous negative economic impact. In 2002, to improve the noise situation in Europe, the EU issued the Environmental Noise Directive (2002/49/EC). The aim of the Directive is to reduce environmental noise and to prevent an increase in noise in areas which are traditionally quiet. For this purpose, noise calculations are conducted for major roads, major railway lines, and various major airports with a high volume of traffic as well as large agglomerations. Until 2018, for this task a national interim method has been used. In 2019, it is replaced by a new calculation procedure, because the European Commission has developed ”Common Noise Assessment Methods in Europe, CNOSSOS-EU”. These methods have been implemented in Annex II to Directive 2002/49/EC in 2015. The transposition of this calculation procedure into German law will be explained in detail.
	Implementation of CNOSSOS-EU method for road noise in Italy Gonzalo De Leon, Francesco Fidecaro, Mauro Cerchiai, Marco Reggiani, Elena Ascari, Gaetano Licitra [Abstract] The European Directive 2015/996 describes the common methods (CNOSSOS-EU) for the determination and management of environmental noise at European level, providing guidelines for a harmonized approach for Member States. CNOSSOS-EU includes different emission data for road, railway and industrial noise and provides a common model for sound propagation. The emission model was derived to be valid under reference conditions in terms of meteorology and traffic as detailed in the reference report by the Joint Research Centre (JRC) of the European Commission. In order to take into account situations that differ from the reference one, appropriate corrections were also offered. For a correct application on a local scale, the default values suggested by CNOSSOS-EU are required to be redefined in order to more adequately describe the Italian reality (i.e. car fleet, surfaces, climate conditions, etc.). Regional Agency for Environmental Protection of Tuscany (ARPAT) carried out specific measurements campaigns to acquire data useful to validate the methods for Italian conditions. At first, regression coefficients used in the emission model, were recalculated from campaigns in 10 different sites carefully chosen to resemble the reference conditions. Finally, different surfaces typologies were studied, with the purpose of obtaining their proper correction factor.
	Matching noise emission from French medium-heavy vehicles and CNOSSOS models Marie-Agnès Pallas, Adrien Le Bellec, David Ecotiere [Abstract] The CNOSSOS-EU prediction method, mandatory since 2019 for noise mapping in compliance with the European Noise Directive, specifies the road vehicle noise equations to be used per vehicle category. French noise emission data differ significantly from the European model and a national decree was published in June 2018 to provide CNOSSOS corrective coefficients in accordance with the national NMPB2008 model. However, the European semi-heavy vehicle category is non-existent in both French emission model and databases, thus lacking to support the proposed coefficients at this stage. To provide missing information, detailed measurements under controlled conditions have been performed on a sample group of semi-heavy vehicles. The contribution of propulsion noise and rolling noise were separated and average emission equations were proposed by subcategory. For two traffic composition scenarios, mainly differing by the ratio of public transport vehicles, a noise emission model has been proposed to represent an average medium-heavy vehicle. The paper presents the approach and the models obtained. The representativeness of the European prediction model in its initial version published in 2015, its French adaptation of June 2018 and the European correction discussed in 2018, is considered for both scenarios. Comparisons of frequency characteristics are emphasized.
	The First Agglomeration Noise Maps in Niš, Serbia, Using CNOSSOS-EU:2015 Simon Shilton, Rafdouglas Tommasi, Jiri Michalik, Álvaro Grilo, Kristina Peric, Ivana Krstic [Abstract] Strategic noise mapping across cities has been undertaken under EC Directive 2002/49/EC since 2006 using a range of national and EC recommended Interim Methods. Under EU Directive 2015/996 the new common methods were introduced which have been the required methods of assessment since 31 December 2018. Within the EuropeAid funded project for strategic noise mapping in City of Nis agglomeration, CNOSSOS- EU:2015 has been used to develop the first city-wide results in the Republic of Serbia. The project has included: the collection of input data from a wide range of National, Regional and City authorities; development of the noise models and noise calculations under CNOSSOS-EU and presentation of the results to the public using a hosted SaaS solution; and verification of the results using long term noise measurement surveys. Based upon the experience gained in Nis, the project is developing guidelines for strategic noise mapping in Serbia using the CNOSSOS-EU:2015 methodology, which will be used as the basis for conducting training of state and municipal authorities in noise mapping in connection with future membership of the EU.

Organization:

S. Luzzi, Vie en.ro.se. Ingegneria M.A. Burgess, UNSW Australia

	Model for total noise assessment under consideration of source specific exposure-response ralationships Manfred Liepert, Johannes Lang, Ulrich Möhler, Dirk Schreckenberg [Abstract] Although most of the German population is exposed to multiple noise sources, noise legislation in Germany is subdivided in regulations for each noise source. In order to improve the quality of life there is a lack of a total noise assessment. The German Federal Environmental Agency therefor initiated the research project ”Model for total noise assessment”. As a part of this project the method of the German VDI standard 3722-2 on the assessment of the impact of noise from multiple sources was further developed. With the help of exposure - response relationships for individual source types a total noise assessment for road traffic, railway, aircraft and industrial noise is possible. For this purpose the exposure-response relationships of the new WHO Environmental Noise Guidelines are used. The possibility of considering health effects and combining annoyance ratings and health effects are discussed. In a second step of this project noise mitigation in case of multiple noise sources are investigated. Different methods for the cost distribution were examined regarding their advantages and disadvantages in the application. The proposed method of cost distribution was finally tested in case studies using the further development of the VDI standard.
	Infrasound and low-frequency noise immission. Structural Vibrations Induced by industrial noise. Improving the ISO 1996- 2:2017 in order to propose a plausible standardized procedure for using in Legal noise assessment Walter Alfredo Montano, Elena Gushiken [Abstract] It is well known that dBA measurements don’t are the best tool to evaluate an alleged noise/vibration disturbance, mainly when impacted dwellings are in a mixed acoustical zones. The results of the measurements, made in Peru inside and outside dwellings subjected to structural vibrations, are presented here. The issue is that there is not always a simple way to distinguish a specific sound from the total sound, and methods recommended in the ISO 1996-2 Standard are very general. An empirical criterion had to be created to find the level of specific sound contained within the total sound. These acoustic determinations consist of removing the outlier values from a time-history filtering of the total sound. In this work, the author proposes a new method of employing statistical tools, adapting Pearson’s criterion, using discrete mathematics, inference, and acoustics criteria in order to develop a computer program to remove spurious noise with not complex algorithms. The purpose of this Paper is to share to the acoustician community that it is plausible to achieve a standardized method, in the sense that an algorithm could be agreed among specialists for having an ”universal computer program” to determine the specific sound contained in the total sound.
	A New Approach for Road Traffic Noise Mapping Using Big Data Gaetano Licitra, Antonino Moro, Luca Teti, Alessandro Del Pizzo, Francesco Bianco [Abstract] The present paper describes a new approach developed within the BEEP project (Big data for Environmental and occupational EPidemiology) aimed to improve road traffic noise mapping in epidemiological studied. The BEEP project, funded by the National Institute for Insurance against Accidents at Work (INAIL), aims at using big data for the evaluation of negative health effects due to air and noise pollution on the Italian general population and the risk of occupational injuries in sub-populations of workers. Noise maps provide noise emissions that are usually calculated by traffic flows measured or derived by a model. The evaluation of traffic flows can determine significant uncertainty in noise estimate: accurate traffic data can significantly improve the meaningfulness of noise models. In this paper, a new method based on Google API and Big Data treatment was developed to estimate traffic flow and produce noise maps of Rome agglomeration. Correlation between travel time and traffic flow considering road characteristics was found using street clustering. Noise maps obtained were compared to those produced by conventional means; results show that the use of big data could improve traffic estimates, in particular during the night period, which is well known to be strongly related to health issues.

Organization:

D. Schreckenberg, ZEUS GmbH T. Yano, Graduate school of Science and Technolody, Kumamoto University

	Noise Annoyance – What does it mean? Paul David Hooper, Ian H. Flindell [Abstract] Current uncertainty regarding to what extent chronic noise annoyance contributes to long term adverse health effects (or not) is not helped by general confusion about what is actually meant by ëchronic noise annoyance’ and the best ways to measure it. The problem arises because the term ënoise annoyance’ means different things to different people in different situations and contexts. The solution to this problem depends on understanding the different situations and contexts in which these terms may be used; understanding the often different requirements of people who use these terms; and reflecting these differences both in the measurement instruments which are actually used and in the constraints and limitations which should be placed on the interpretation of those measurements in different situations and contexts. In the paper we provide practical examples to illustrate how this problem arises and how it can be solved in real-life situations and contexts.
	Measuring Noise Annoyance with Multiple Question Scales - an Overview Jördis Wothge [Abstract] Noise annoyance is a diverse concept. This is reflected by the great variety of different definitions and understandings that exist in the noise impact research community. Noise annoyance can be defined as an emotion, a rational decision, knowledge, an attitude or the result of a disturbance. Noise annoyance can also be regarded as a multi-dimensional construct that entails some or all of the aspects mentioned above. Currently, most studies on noise annoyance make use of the widely accepted single-item ICBEN question to measure noise annoyance. There is, however, an ongoing discussion in the research field whether a single-item question can comprehensively cover all facets of noise annoyance. This lead to the postulation of multi-item questionnaires to measure noise annoyance in order to reflect the multi-dimensionality of noise annoyance more adequately. The current contribution aims to give an overview about different multiple question scale approaches to measure noise annoyance.
	Assessment of the impact of changes in noise exposure at an expanding airport by means of the multiple item aircraft noise annoyance scale (MIAS) Dirk Schreckenberg, Rainer Guski, Julia Haubrich, Jan Spilski [Abstract] The Multiple Item Noise Annoyance Scale (MIAS) measures noise annoyance as a multidimensional concept including 1) the experience of an often repeated noise-related disturbance and the behavioural response to it, (2) an emotional/attitudinal response to the sound and its disturbing impact, (3) the perception of control of the noise situation. MIAS consists of seven items including the 5-point ICBEN noise annoyance scale and the sub-dimensions ënoise disturbances’ and ëlack of coping capacity’, each consisting of three items. Its psychometric properties are investigated for aircraft, road, and railway noise annoyance. In this study, we analyse changes in values of the sub-dimension of MIAS for aircraft noise annoyance before and after the opening of a new runway and the implementation of a night curfew in communities around Frankfurt Airport. Previous studies have shown evidence of a shift in exposure-response relationships for annoyance in situations of changes in noise exposure (change effect). With the assessment of MIAS and its components it can be identified, whether the change effect is more pronounced in the ëdisturbance’ component of annoyance or in the lack of capacity to cope with the (new) noise situation. This improves the understanding of annoyance in situations of changes in noise exposure.
	Forty-five years of surveys on annoyance from road traffic noise Truls Gjestland [Abstract] The Miedema & Vos curve for predicting road traffic noise annoyance was based on 18 different surveys conducted over a period of 9 years. The curve was developed using a statistical regression technique. An extended analysis comprising 46 surveys conducted between 1969 and 2013 yields a very similar dose-response relationship. This analysis was based on the CTL method (Community Tolerance Level). There are no obvious temporal trends, and contrary to claims by other researcher the prevalence of highly annoyed residents exposed to road traffic noise today is the same as 45 years ago.
	Laboratory Studies of Annoyance Sonoko Kuwano, Seiichiro Namba [Abstract] It is said that in the concept of annoyance non-acoustic factors are included as well as physical factors of sounds. Therefore, it is difficult to investigate annoyance in laboratory studies where the situation is somewhat different from daily life situations. Annoyance is usually asked in social surveys. Laboratory studies have merits that various parameters are precisely controlled The authors tried various laboratory studies of annoyance by developing and controlling the experimental methods. One of them is the method called ”subject-interrupted noise source method”. The study using this new method and other laboratory studies of annoyance will be introduced in this paper.
	Perception of low-level Sound Sources in Everyday Situations Siegbert Versümer, Jörg Becker-Schweitzer [Abstract] Noise research has focused on urban sounds like traffic or machine noise. The present paper, in contrast, investigates low-level sound sources that can affect human wellbeing and health and that are only perceived in specific situations. To address our research questions, we conducted an online survey with around 2000 participants representative for the German population. In the course of the survey, participants reported on situations in which they typically noticed low-level sound sources. They described the characteristics of the sounds as well as their frequency of occurrence. In addition, they reported on typical activities in these specific situations and personal strategies to cope with negative effects. Finally, they filled out questionnaires measuring the Big Five personality traits extraversion and neuroticism as well as noise sensitivity. The results include a ranked list of sound sources that people classify as being ”low-noise” and that show a potential to affect humans in a negative way. They further constitute the basis for a multi-level prediction model on the effects of low-level sound sources including situational and person-related factors. Therefore, the findings contribute to a deeper understanding and to an improvement of low-level acoustic environments.
	Community response to high-speed railway noise in Tianjin, China Lan Zhang, Hui Ma [Abstract] Adverse effects of noise from high-speed railway have become more and more serious and widespread with the rapid development of high-speed rail in China. A social survey on community response to high-speed railway noise was carried out in Tianjin, China from March to July, 2018 along the Beijing-Tianjin intercity high-speed rail line and Beijing-Shanghai high-speed rail line. 674 residents were interviewed by questionnaires to explore their noise annoyance, activity interference caused by high- speed railway noise, sensitivity to noise, and satisfaction with living environment. The exposure of high-speed railway noise was measured for 24h at each site, and standardized questions and scales suggested by ICBEN Team 6 were used in the social survey. As a result, the dose-response relationship curve for high- speed railway noise of Chinese residents was established, and it showed that with the increasing of noise level (Ldn), annoyance increased exponentially. Compared with Japanese and French people, Chinese residents showed more annoyed to high- speed railway noise. It also suggested background noise had important impact on noise annoyance evaluation to high-speed railway. Furthermore, annoyance caused by high-speed railway noise was also greatly influenced by subjective characteristics such as noise sensitivity, satisfaction of living environment and attitudes.
	Using mobile application to assess quality of acoustic and visual environment in relationship with aircraft noise Ferenc Marki, Catherine Lavandier, Dirk Schreckenberg, Stephan Grossarth [Abstract] Aircraft noise annoyance has been studied for more than 50 years, asking people to rate their annoyance on a unidimensional scale. The importance of non-acoustic factors shows that a more global approach could help to understand how people perceive their acoustic environment. Inspired by the soundscape studies, and in the frame of the ANIMA European project on the impact of aviation noise, an Experience Sampling Method with signal-contingent sampling is chosen to question people in their daily live about the quality of their acoustic and visual environment, and not on the annoyance. A dedicated mobile application is developed, which collects data each hour of a day (night periods are excluded).The experiment lasts two or three weeks depending on the number of notifications per day the participant found acceptable. After the notification, participants are asked to record acoustic measurements (third octave bands, each second during one minute). After the questionnaire, participants are asked to take a picture of the environment. Summary questionnaires are filled at the end of each week, and a global questionnaire is filled at the end of the experiment. This paper focuses on a pilot study whose aim is to evaluate the acceptability of the method by participants.
	Assessment of Short-term Annoyance due to Shooting Noise Using the Experience Sampling Method Stephan Grossarth, Dirk Schreckenberg [Abstract] Experience Sampling or Ecological Momentary Assessment allows researchers to assess behavioural and psychological phenomena directly in the situation of occurrence/in a given situation. In a longitudinal study, we aim to model short term annoyance due to shooting noise as a function of intra- and inter-individual differences in perceived annoyance by using a hierarchical data modelling approach. The MovisensXS app is used to collect data from three different study sites in Germany from 80 Participants. They are asked to participate within two weeks for a period of five consecutive days per week and, during this time, to fill out a survey, consisting of 19 items, five times a day at set intervals. This will add up to approximately 4,000 single measurements, depending on the response rate and the compliance. We describe the methodology of the study, the planning and executional steps taken to ensure a proper data quality. Furthermore, we will explain what has been done to prepare for the data analysis. Finally, we will give a forecast into data analysis and will exemplarily talk about longitudinal data analysis and linear mixed effect modeling.

Organization:

A.L. Brown, Griffith School of Environment, Griffith University, Australia T. Van Renterghem, Ghent University, Department of Information Technology, WAVES research group

	Comparing noise policies of 8 European cities using a noise intervention classification scheme Timothy Van Renterghem, Hannelore Hernalsteen, Alan Lex Brown [Abstract] Over the last decades, cities have been developing noise policies. Comparing such efforts, however, is not a simple task. For this purpose, a previously proposed noise intervention classification scheme is used as a framework. This framework discriminates between source interventions (type A), path interventions (B), closing/opening new infrastructure (C), other physical interventions (D, e.g. achieving a quiet side) and communication/education measures (E). Responsible officers dealing with noise at environmental departments of 8 large European cities were interviewed. The invitations for an interview were sent out taking into account the geographical distribution of cities over Europe while a reasonably high position on a sustainable cities ranking was required. The interviews focused on the noise problem definition and what is in the noise management toolbox of the city. In addition, it is assessed how noise policies are evaluated, and to what extent public participation in the noise policy and planning process occurs. The interviews show highly contrasting opinions and practices among the interviewed cities at almost any aspect of the noise management process.
	Are perceived noise control and its value related to behavioural determinants of residents’ civic engagement? A cross-sectional study among older adults Natalie Riedel, Adrian Loerbroks, Joachim Scheiner, Susanne Moebus, Irene Van Kamp, Thomas Claßen, Heike Koeckler, Gabriele Bolte [Abstract] We had suggested that residents’ perception and value of being able to control one’s personal traffic noise exposure at home are relevant for the readiness to participate in noise policies. Further, we had framed residents’ participation readiness by specifying behavioural determinants (positive attitude towards and subjective norm of taking action for a quiet environment as well as behavioural control perceptions, i.e. noise-related institutional knowledge, self-efficacy, and communal mastery). Using cross-sectional questionnaire data on about 1,600 older adults in Germany in 2016 and logistic regression analyses, we modelled the probability of agreement to each behavioural determinant in relation to residents’ perceived noise control (PNC) and its value as measured by six point Likert-scales. Besides other covariates, knowledge of public consultation in noise action planning, internet access, and social network were added to the models. Agreement varied from 36% for a positive subjective norm to 72% for a positive attitude. Overall, results confirmed positive associations between PNC, its value and behavioural determinants (e.g. ORPNC= 1.21, 95%-CI 1.09-1.35, ORvalue=1.20, 95%-CI 1.13 1.28 for behaviour-specific self-efficacy), indicating a link between perceived exposure-specific characteristics and participation readiness. Next, we will investigate whether our selected behavioural determinants statistically predict intended and performed civic engagement.
	Soundscape approaches in urban planning: implications for an intervention framework Irene Van Kamp, Alan Lex Brown, Dirk Schreckenberg [Abstract] There is increasing demand for innovative approaches to decrease the negative impact of noise by sound-conscious design. While the health effects of environmental noise are well mapped, little is known about the effectivity of interventions on health. This is one of our conclusions in the WHO-review on health impacts of noise interventions. More and better interventions studies are needed, in particular for sources other than road traffic, and for health outcomes other than annoyance. Soundscape interventions were in this review classified under the category” Other physical interventions”. Three out of four studies in this domain demonstrated potential efficacy. Although soundscapes seem essential for urban planning, guidelines on how to implement this knowledge are missing. Against this background an inventory of architectural soundscape projects in the Netherlands and elsewhere was made, combining interviews with experts and planners with a literature review. This led to the conclusion that well documented approaches all combined changes in the physical environment, with stakeholder involvement. In that way a sense of ownership was created by the interventions, forming an additional category in the typology of interventions as canvassed earlier. This paper suggests some extensions the typology of interventions.
	Road Traffic Noise Interventions: Development of a Method to Quantify their Effects on Annoyance and Sleep Disturbance on a Small Urban Scale Maud Dohmen, Maarten Hornikx, Irene Van Kamp [Abstract] Road traffic is the main source contributing to noise exposure in urbanized areas. With the rapid urbanization and the increasing attention on the way sound can affect health and well-being, the question is how to shield ourselves from harmful effects. The most common noise intervention research studies regard only large-scale interventions, like noise barriers along a highway, the effect on health and well-being are often based on rough estimations of sound reduction. In this study, the effect of a small scale intervention on annoyance and sleep disturbance will be investigated, considering relative changes in dose-response relationships to account to local effects. It will give more insight into the subtle changes due to interventions on a local scale, providing more detail on top of existing conventional studies. To develop this methodology an intervention study in Rotterdam is taken as a test case. The location and its intervention are modeled using 3D-geometrical acoustic software (Olive Tree Lab). The predicted noise reduction due to the interventions, as well as the predicted health effects are compared with measured data and local surveys. The results shed a light on the suitability of this approach to predict health effects due to interventions at a local scale.

Organization:

S. Yokoyama, Kobayasi Institute of Physical Research F. Van Den Berg, University of Groningen

	How do audio and visual characteristics of wind turbines contribute to noise annoyance? Beat Schäffer, Reto Pieren, Ulrike Wissen Hayek, Nadine Biver, Adrienne Grêt-Regamey [Abstract] Wind farms often evoke strong annoyance reactions in residents. Literature suggests that both acoustical characteristics and the visibility of wind turbines may contribute to noise annoyance. However, studies on the mutual audio-visual effects on annoyance are still rare. The objective of this study was to investigate the short-term noise annoyance reactions to different wind turbine noise situations in a controlled laboratory experiment. A set of 24 audio-visual rural scenarios containing a single wind turbine was synthesized, i.e., visualized and auralized. Combined with the full factorial design of the experiment, this allowed separating the individual contributions of the following variables to noise annoyance: distance to the wind turbine, periodic amplitude modulation of the sound (with, without) and visual setting (landscape with visible wind turbine, landscape only, grey background). The experiment revealed that both visual and acoustical characteristics strongly affect noise annoyance. In particular, annoyance increased with periodic amplitude modulation and/or when the wind turbine was visible, but decreased when only the landscape was visible. While the acoustical effects could be reliably assessed, the visual effects were less straightforward to reveal, as they are afflicted with carryover effects. The presentation order of audio-visual stimuli may therefore be crucial for study outcomes.
	Implementation Of The Issue Of Noise From Wind Turbines At Low Frequencies Martino Marini, Costantino Carlo Mastino, Roberto Baccoli, Andrea Frattolillo, Antonino Di Bella [Abstract] The enduring energy scenario leads to further promote the development of the exploitation of renewable energy sources. Recent European standards, have defining a path to reach in 2050 a level of decarbonization lower of 80% compared to 1990. Wind farms have been growing quickly for last decade with individual wind turbines getting larger and larger. In addition to the above benefits drawbacks has arisen, due to noise generation from wind turbines and adverse reaction of some nearby residents. The noise generated by wind turbines has a broad spectrum character but the low frequency noise causes special problems.It is a fact that in different European countries special laws have been adopted to impose noise limits and evaluation methods for the assessment of environmental low frequency noise from this kind of sound sources. Other countries are still lacking specific rules but in the authorization procedure such analysis is required by environmental control agencies. The purpose of this study consists of comparing the assessment procedures currently used in different European countries for the prediction of low frequency noise from wind turbines and its propagation. The comparison of procedures gives a chance to put forward progressions in low frequency noise emission and reception.
	Pilot study on perceived sleep acceptability of low-frequency, amplitude modulated tonal noise Kristy Lee Hansen, Phuc Nguyen, Branko Zajamsek, Gorica Micic, Peter Catcheside [Abstract] The presence of amplitude modulation (AM) in wind farm noise has been shown to result in increased annoyance. An equivalent unmodulated noise would have to be louder by several decibels to be as equally annoying as modulated noise. Despite this finding, many standards and guidelines do not specify a penalty for the presence of AM. This study investigates the prevalence of AM over a 1 year period at four different residences located between approximately 1 km and 3.5 km from three different wind farms. The analysis is based on continuous measurements of raw time data that were collected using remote monitoring systems. Preliminary results will be presented based on the data that has been collected and analysed to date. As well as considering the percentage of time that wind farm AM is present, this investigation considers other factors such as the AM depth and wind farm power output. The results provide insight into the prevalence and characteristics of AM at various distances from a wind farm over a significant period of time.
	Epidemiological study on long-term health effects of low-frequency noise produced by wind power stations in Japan Tatsuya Ishitake, Yoshitaka Morimatsu, Kunio Hara [Abstract] We investigated whether long-term exposure to wind turbine noise (WTN) including low-frequency noise generated by wind power facilities is a risk factor of sleep disorders. We performed an epidemiological study of living environment and health effects, surveying 9,000 residents (≥20 years) living in areas with operational wind power facilities. Sleep disorders were assessed using the Athens Insomnia Scale. To assess environmental noise in residential areas near the wind turbines, low-frequency sound exposure levels were measured at 50 community centers of the town. Multiple logistic regression analysis was used for evaluation of a risk factor for several noise exposure indices. Significant relationships between the distance from the nearest WT to dwellings and hearing, annoyance, sleep disorders were observed. By multiple logistic analysis the prevalence rate of sleep disorders was significantly higher for residents who reported subjectively hearing noise being than for those who did not. Moreover, the reported prevalence rate of sleep disorders was significantly higher in residents living at a distance of ≤1,500 m from the nearest wind turbine compared to that for residents living at a distance ≥2,000 m. The attitudes of residents towards wind power facilities and sensitivity to noise strongly affected their responses regarding sleep disorder prevalence.
	Perception of low-frequency components contained in general environmental noises including wind turbines Sakae Yokoyama, Tomohiro Kobayashi [Abstract] The influence of low frequency components in noises including wind turbine noise has become a serious problem and various researches are being made in many countries. Also in Japan, synthetic study programs mainly targeted for wind turbine noise have been performed, in which field measurements, social survey on the response of nearby residents and laboratory experiments on audibility of low frequency sounds were conducted. In parallel to these studies, field measurements on general environmental noises including transportation noises inside and outside of various vehicles were performed by paying attention to low frequency components. As a result, it has been found that low frequency components are included not only in wind turbine noise but also in general environmental noises. The environmental sounds recorded on sites were reproduced by a test facility which can reproduce low frequency components including infrasound and auditory tests were performed. As a result, it has been found that the A-weighted sound pressure level is robustly applicable to the assessment of loudness for such kinds of environmental sounds and the low frequency components contained in wind turbine noise around residential area is not audible.
	How to present pure-tone infrasound to the ear Holger Joost, Björn Friedrich, Jesko Verhey, Thomas Fedtke [Abstract] An increasing number of individuals are being exposed to infrasound. It is, however, still not clear, how infrasound is perceived by human beings. Regarding auditory perception, one hypothesis is that non-linear processes within the ear generate distortion products that make infrasound audible. To investigate this hypothesis, it is crucial to rule out that such distortions were produced technically, i.e., by the stimulus presentation system rather than the auditory system. This paper introduces a compact, low-distortion infrasound system, based on a commercially available audiometric earphone, for simultaneous presentation of pure-tone infrasound and audio sound to the ear via an ear insert. This system also allows for the recording of sound in the ear canal during signal presentation. Performance characteristics of the system for the stimulus presentation and the sound recording units are presented. It is shown that the interaction of infrasound and audio sound, such as modulation of the audio sound by the infrasound, is negligible with respect to auditory perception.
	Frequency characteristics of oppressive and vibratory feeling to low-frequency sound Makoto Morinaga, Ippei Yamamoto, Tomohiro Kobayashi, Koichi Makino, Hiroaki Ochiai, Hideki Tachibana [Abstract] Although the threshold for oppressive and vibratory feeling for low frequency sound are widely known, the experiments based thereon were conducted about 40 years ago in Japan, and there are some questions to be examined. For example, although the term ”oppressive and vibratory feeling” is generally unfamiliar words, it is unknown how the instruction was provided to the participants about the meaning of the term. In the present study, an experiment using similar method to the previous study was conducted, and the threshold of the feeling was reviewed. Using one-third octave band noises with the center frequency in the range from 10 Hz to 630 Hz, laboratory experiment was conducted to investigate the value of sound pressure level where oppressive and vibratory feeling preferentially appeared for each frequency. Participants were divided into two groups; one is the group provided instruction for the meaning of oppressive and vibratory feeling and the other is without such detailed instruction, and the results were compared. It was suggested that the sound pressure level where the feeling preferentially occurs is slightly different between the two groups. Furthermore, it was found that the feeling appeared even in the higher frequency band compared with the previous study.
	The effect of brown and black noise on persons suffering from a low frequency sound Frits Van Den Berg, Karin De Boer [Abstract] Some people suffer from hearing an often untraceable low-pitched sound. To help them two types of low frequency noise (brown and black noise) were put on CD. Sufferers could obtain this free CD on request. Over the course of more than 10 years about 200 CD’s were sent out. In these 10 years 26 applicants have given a spontaneous reaction by e-mail after having used the CD for a shorter or longer period of time. 23 of these comments were (very) positive, 2 comments stated the CD did not help. In 2018 a retrospective survey was designed for those that had requested a CD before November 2018. The main goal of this study is to investigate how the complaints have changed over time in relation to the use of the CD. Further questions examine which other measures participants have taken and how they use(d) the CD in practical terms (brown and/or black noise, playing time, frequency of use, volume). The study will start in February 2019 and a first analysis will be given in the paper. This may inspire others to help sufferers by using the same method. This could be an opportunity to include more participants in different areas.
	Equal annoyance contours at frequencies 4 - 1000 Hz Ville Rajala, Jarkko Hakala, Valtteri Hongisto [Abstract] BACKGROUND. Knowledge of human perception of infrasound is very limited. Only few publications deal with the hearing threshold and subjective loudness of infrasound. Our aim is to create equal loudness contours extending to infrasound frequencies and to examine hearing threshold of infrasound by using sine tones in frequency range starting from 4 Hz. METHODS. A psychoacoustical experiment involved 20 participants in the 19-26 years age group. The experimental sounds were sinusoidal tones in frequency region 4-8000 Hz. The participant was located in a pressure chamber. Low frequency tones were played by using a special loudspeaker system. High frequency tones were played with headphones. The hearing threshold was determined by using staircase method. Loudness levels 20, 40 and 60 phon were defined by using randomized maximum likelihood sequential procedure. The equal loudness contours will be calculated by using methods described in ISO 226. RESULTS. The experiment is ongoing and the results will be available until the deadline of the full paper (1 June, 2019).
	Portable Infrasound Monitoring Device with Multiple MEMS Pressure Sensors Ryouichi Nishimura, Zhenglie Chi, Yôiti Suzuki [Abstract] MEMS sensors are mass-product and therefore expected to be available to produce a low cost device for infrasound monitoring. A low cost device is useful in constructing a sensor network with various scales from a small to large number of sensors. To test its feasibility, we developed a device consisting of 32 MEMS pressure sensors. A laboratory experiment showed that internal noise increases as signal frequency goes high but the noise can be reduced by simple averaging. Based on this fact, we stacked eight MEMS pressure sensors on a single board to assemble an infrasound monitoring device which enable to detect a signal with amplitude of more than approximately 5 Pa. A field test was carried out, deploying them in the surrounding area of an active volcano, Sakurajima in Japan. Signals associated with explosive eruptions were clearly observed in the frequency region between 0.1 Hz to 5 Hz. The source position was also estimated using the signals observed at four different points. The estimated source position showed in general a good agreement with that of the crater of the volcano. It was also implied that signals around the acoustic cut-off frequency can also be observed.
	Infrasound research activities in Kochi University of Technology - Infrasound observation network by using multiple comprehensive infrasound sensors and its application for disaster mitigation as well as safe human life - Masa-Yuki Yamamoto [Abstract] Infrasound research activities in Kochi University of Technology will be presented. Infrasound observation network by using multiple comprehensive-type infrasound sensors was established recently in Kochi, Japan. Main purpose of the sensing network is for disaster mitigation/prevention mainly for tsunami, indicating a good application of infrasound sensing. If we can detect infrasonic waves before coming real tsunami waves to sea shore, it can effectively be used for saving human lives from the flood of sea water. However, the sensors can also detect many types of natural and artificial low-frequency audible sound and infrasound. It means the network sensing system can be applied to safe human life by means of emergency level alert to local citizens like weather forecasts. Moreover, we recently tried to evaluate wind turbine low-frequency audible/infrasonic noise affection for human health. In a vicinity of two wind turbines we measured sound field environment by using audible/infrasonic sensors together with biological information sensing system with multiple methods with respect to more than 30 human subjects/testers. With recent progress of biological information sensor technology, we can use them freely outside but many kinds of artifacts can also affect real signals of heart beat and brain wave measurements like our noise hearings.

Organization:

R. Guski, Ruhr-University Bochum M. Morinaga, DFEIA

	Short-term annoyance from nocturnal aircraft noise in children: The influence of acoustical and non-acoustical factors Julia Quehl, Susanne Bartels, Rolf Fimmers, Daniel Aeschbach [Abstract] Acute psychological effects of aircraft noise have mostly been investigated in healthy adults so far. Recent noise effects research is increasingly focusing on individuals particularly sensitive to harmful noise effects. These vulnerable groups include especially children. Almost no studies exist on effects of nocturnal aircraft noise on short-term annoyance response in children. We aimed at closing this research gap by means of a field study conducted around Cologne-Bonn Airport in 51 children aged 8 to 10 years. Children were examined during four consecutive nights. During all nights, aircraft noise exposure was recorded inside the children‘s bedroom. Aircraft noise metrics were calculated based on these measurements. Participants retrospectively rated short-term annoyance each morning via a five-point scale. Information on relevant moderating psychological factors (e.g., perception and evaluation of air traffic in the residential area, personal variables including sensitivity to noise) was obtained in personal interviews on the first study day. Logistic regression analysis was used to develop new exposure-response curves as a function of different acoustical and non-acoustical variables. Unlike adults, children’s short-term annoyance was not influenced by noise metrics such as number of nocturnal aircraft noise events. Noise sensitivity and attitudes towards air traffic, however, had a significant effect.
	Noise-induced Annoyance due to Nocturnal Road Traffic: Results of a Field Study Sarah Weidenfeld, Eva-Maria Elmenhorst, Sarah Sanok, Uwe Müller, Daniel Aeschbach [Abstract] Since many traffic noise studies are based on laboratory surveys or exhibit shortcomings in noise measurement, there is a lack of valid exposure-response relationships between traffic noise and annoyance. To fill this gap for road traffic, precise measurements of noise parameters (e.g. energy equivalent sound pressure level [LAeq]) are needed. In this field study 40 healthy participants (mean age = 29.1; SD = 11.7; 26 females) rated their annoyance due to moderate road traffic noise exposure during the previous night on four consecutive mornings. Further questionnaires were used to measure psychological, sleep-related and subject-related variables. Measurements of acoustic parameters were undertaken inside the residents’ bedroom. A generalized estimating equation model (GEE) revealed a statistically significant rise in the annoyance with increasing LAeq of the traffic noise (p = .023). The extent of the subjectively perceived noise exposure had a reinforcing effect (p < .001), whereas a high subjective sleep quality had a reducing effect (p = .007) on annoyance. The present study emphasizes the importance of sleep in annoyance responses and delivers a prediction model for noise-induced nocturnal short-term annoyance in areas with moderate road traffic. Prospective investigations should be extended to residential areas with dense road traffic.
	Effects of The Number of Noise Events from SHINKANSEN Railway on Annoyance Keiji Yagawa, Ayumi Shiotani, Toshihiko Matsui, Takashi Yano [Abstract] This paper describes the relation between the number of noise events of Shinkansen and the increasing noise annoyance. In Japan, the maximum noise level (LA,Smax) is adopted as the noise index for the environmental quality standards for Shinkansen railway noise since 1975. As the number of trains has increased with the development of Shinkansen railway network, it has been debated which of LA,Smax or energy averaged indices such as LAeq,24h and Lden is more appropriate for the index. The Ministry of Environment initiated nation-wide social surveys on Shinkansen noise in 2014 and 2017 in the areas along Shinkansen railways which had 40 to 340 Shinkansen service a day. The total sample size was 1455 and LA,Smax ranged from 53 to 84 dB. We hypothesized that annoyance for Shinkansen railway noise increased according to the increase of noise events even if maximum noise levels are equal. It was confirmed that the number of noise events increased annoyance for Shinkansen railway noise up to around 200/day, but the effect saturated in the range more than 300/day. Comparing logistic regression models between high annoyance and LA,Smax/Lden, either of LA,Smax or Lden is not always superior to the other.
	Leq + X: Re-Assessment of exposure-response relationships for aircraft noise annoyance and disturbances to improve explained variance Julia Haubrich, Mark Brink, Rainer Guski, Ullrich Isermann, Beat Schäffer, Rainer Schmid, Dirk Schreckenberg, Jean Marc Wunderli [Abstract] There is evidence from literature that an increased fraction of people is highly annoyed by aircraft noise by given average noise levels than in previous decades. Among potential reasons are changes of the air traffic over the last decades such as the increased number of movements along with a less noisy fleet mix that might be relevant for noise responses but are not adequately reflected by average noise metrics. In the Leq+X project, data of two German and two Swiss field studies comprising aircraft noise exposure and annoyance data of about 37’700 residents living around altogether seven airports are re-analysed. Among others, the analyses include logistic regression models of the percentage (highly) annoyed, (highly) sleep disturbed and (highly) disturbed in communication on the Leq (day/night), on the Leq combined with either maximum sound level (Lmax) or number of events above threshold (NAT), and on the Lmax combined with NAT for different thresholds. Further models include the Leqk with non-energy-equivalent parameters k = 20 or 30, which weight the number of events more strongly than the energy-equivalent Leq (k = 10). In this paper, the methodological approach and first results will be presented.
	Do we need different aircraft noise metrics to predict annoyance for different groups of people? Jan Spilski, Kirstin Bergström, Ulrich Möhler, Thomas Lachmann, Maria Klatte [Abstract] In the literature, LAeq and Lden are the dominating noise exposure metrics for calculating the relationship between noise and annoyance. However, alternative metrics such as the number of events above a threshold (NAT), LAmax or Emergence are rarely considered, although they may have the potential to explain additional variance. This seems necessary because the average relationship between aircraft noise levels and annoyance raw scores is .44 (Guski et al., 2017) which represents an explained variance of only 19.36% (R2 = .1936). Here, we present a post hoc analysis of the NORAH data set (Klatte et al., 2016) and aim to evaluate the adequacy of alternative noise metrics. We calculated multilevel analyses with robust estimation algorithms (MLR) and using different noise metrics (LAeq, LAmax, Emergence, NAT). The analyses were carried out for different groups of people (children, parents, teachers) and different contexts (school: workplace or learning environment, residential environment). The calculations with the NAT60 criterion showed an increase of explained variance (R2) in peopleës annoyance. The effect was robust for different contexts and groups of people.

Organization:

S. Yokoshima, Kanagawa Environmental Research Center O. Breugelmans, Netherlands National Institute for Public Health and the Environment

	Combined effect of vibrations on railway noise annoyance Shigenori Yokoshima, Takashi Morihara, Yasunao Matsumoto [Abstract] In Japan, noise annoyance associated with the Shinkansen railway is more serious than that associated with conventional railway, as Yokoshima at al. reported in the previous study. The reasons for the difference have been explored in psychological and physical aspects. Tamura pointed out in a comparison survey along the Shinkansen and conventional railways that people living along the Shinkansen railways had negative attitudes to noise source. Therefore, no recognition of necessity of the Shinkansen railway noise brought about higher annoyance than that of conventional railway noise. Following this study, Yokoshima and Tamura clarified the combined effect of vibrations on noise annoyance associated with the Shinkansen railway trains. However, the combined effect of vibrations on noise annoyance from conventional railway trains, including the comparison of the Shinkansen railway trains, has not quantified. We performed a secondary analysis using individual datasets equipped with noise and vibration exposure as well as noise annoyance associated with the railway trains. Applying a multiple logistic regression analysis to the datasets, we clarified individual combined effect of vibrations on noise annoyance associated with the Shinkansen and conventional railways and discussed the impact of combined effect on the prevalence of noise annoyance.
	Experimental Investigation of Evaluation Method of Horizontal Vibration in Building Caused by External Vibration Sources Kentaro Hayashi, Yasunao Matsumoto, Toyohiko Higashida [Abstract] The purpose of this study was to examine the evaluation method based on subjective responses to vibration caused by external vibration source including road traffic and railroads for building habitability. The target buildings were small-scale buildings, such as detached houses in Japan whose main structure was wooden or steel frame. The first natural frequency of vibration mode of those building which has dominant horizontal vibration has been known to be in the range from 4Hz to 8Hz. Horizontal vibrations at the ground in that frequency range can be amplified due to the dynamic characteristics of buildings. An experiment involving human subjects were conducted to investigate relationship between subjective responses and evaluation value computed from vibration acceleration time histories. The input stimuli consisted of sinusoidal vibrations at 4Hz and 8Hz for several different durations and vibrations based on building vibrations measured in actual residential buildings. 7- step evaluation scales composed of expressions to provide equal psychological intensity intervals about ”Magnitude”, ”Concern” and ”Discomfort” were used to measure subjective responses in the experiment. It was found that Vibration Dose Value(VDV), which evaluated the effect of vibration duration by the fourth power, was most appropriate among the evaluation methods used in this study.
	Case examples of the measurement of vibration and annoyance response in residential buildings beside roads and railways Yasunao Matsumoto, Shigenori Yokoshima, Kentaro Hayashi [Abstract] Vibrations induced by road traffic and trains often degrade the habitability of surrounding buildings. The Vibration Regulation Law was enforced in 1976 in Japan. Road traffic vibration is a target of regulation, although there have been many cases in which the assessment based on the law is inconsistent with adverse comments on vibration raised from building occupant. An investigation was conducted to understand the relationship between vibrations inside and outside of building and annoyance response of building residents. Vibration measurements were made at several residential buildings located beside roads and railways. Those buildings were single- family houses, which were two- or three-story wooden building. Acceleration time histories were recorded at three locations inside and outside of buildings to understand the dynamic characteristics of those buildings. The buildings for measurement were selected from houses of respondents in a separate questionnaire survey on vibration annoyance. Annoyance responses estimated from measured vibration records with the application of different metrics available in current standards and guidelines were compared with the annoyance responses determined in the questionnaire survey. Subjective responses that were estimated based on the results of separate laboratory experimental study were also compared with the annoyance responses in the questionnaire survey.
	Annoyance reactions due to noise and vibrations caused by different train types in Sweden: results from the EpiVib study Elise Van Kempen, Mikael Ögren, Laura Maclachlan, Laith Hussain-Alkatheeb, Kerstin Persson Waye [Abstract] The few existing exposure-response relationships describing the association between exposure to vibrations from rail traffic and annoyance were not always able to distinguish between the different types of trains. However, indications were found that exposure to vibrations caused by freight trains are stronger related to annoyance than exposure to vibrations caused by passenger trains. Although the results of the studies looking into the difference in exposure-response functions describing the association between exposure to noise from trains and annoyance between the different train types are not consistent, there are also concerns that noise from freight trains may be more annoying than noise from passenger trains. This paper reports on a study investigating annoyance reactions due to vibrations and noise from different types of trains. To this end, we used data from the Swedish EpiVib project, collected by means of a survey carried out in 2017 among more than 7,200 people aged 18-80 years living in the Västa Götaland, Värmland and Örebro regions in Sweden close to the railway track.
	Effects of changes in operational and residential factors on public health and reactions at the vicinity of Noi Bai International Airport Thulan Nguyen, Bach Lien Trieu, Takashi Yano, Takashi Morihara, Yasuhiro Hiraguri, Makoto Morinaga [Abstract] The rapid development of air transport, together with the urbanization of the area around the airports, resulted in increased levels and wider influential scale of noise due to the airport operation. The number of flights operated and the population density at residential areas around Noi Bai International Airport has increased significantly in the past nine years. The exposure noise levels (Lden) at surveyed sites ranged between 48 and 61 dB in 2009, but increased to range between 54 and 76 dB in 2017. General health indicators such as Body Mass Index (BMI), blood pressure, etc. were investigated in two surveys conducted in November 2017 and August 2018. Exposure-response relationships established in the five surveys conducted from 2014 to 2018 were found to be higher than that established in 2009 survey. Comparisons of respondents with high blood pressure and insomnia ratios at different noise exposure level ranges showed that there is no significant association between ratios of hypertension and noise exposure levels (Lden) but a significant exposure-response relationship was found between night-time noise exposure levels and insomnia. This result suggested that a restriction on nighttime flight operation should be considered to protect health of resident living around airports in Vietnam.

Organization:

M. Basner, University of Pennsylvania M. Smith, University of Pennsylvania Perelman School of Medicine

	Pilot Field Study on the Effects of Aircraft Noise on Sleep Around Atlanta International Airport Mathias Basner, Michael Smith, Sarah Rocha, Maryam Witte [Abstract] Aircraft noise can disrupt sleep and impair recuperation. The most recent US investigation on the effects of aircraft noise on sleep was conducted more than 20 years ago. Since then, traffic patterns and noise levels produced by single aircraft have changed substantially. It is therefore important that field studies be conducted in the US. to acquire current data on sleep disturbance relative to varying degrees of noise exposure. To inform the design of a larger national study, a pilot field study was conducted around Atlanta Airport. The primary goal was to assess feasibility of acquiring acoustical and physiological data with no investigator on site and equipment mailed to participants who then applied electrodes and started and stopped measurements each night on their own. We mailed 4080 recruitment questionnaires to randomly selected households exposed to aircraft noise (>35 dB LAEq,23- 07,outdoor). Among the 407 respondents, 34 participated in the unattended five night in-home study. Indoor sound levels were determined via measurement. Arousal from sleep was recorded via a device measuring body movement and electrophysiological heart rate. Self-reported sleep was assessed with morning questionnaires. Several ways to improve data quantity and quality were identified. Overall, the approach was found to be feasible.
	Effects of nocturnal aircraft noise on objective and subjective sleep quality in primary school children Susanne Bartels, Julia Quehl, Daniel Aeschbach [Abstract] Little is known about childhood sleep exposed to transportation noise. We investigated the effect of aircraft noise on sleep in a field study in 51 children aged 8 to 10 years who resided near a 24-h operating airport. Sleep was measured polysomnographically during four nights. Besides, subjects rated their sleep quality and current fatigue each morning. Aircraft noise exposure was operationalized by the number of aircraft noise events above 30 dB(A) per night derived from continuous acoustic recordings at the children’s ear. For the prediction of objective and subjective sleep quality, we applied mixed models with random intercept and the number of noise events as dichotomous factor (median split at 37 events). Mixed models were also used to analyze the relationship between objective and subjective measurements. Noise affected the macrostructure of sleep. A higher number of noise events was associated with a reduction of slow wave sleep and an increase of wake time after sleep onset. In contrast, aircraft noise exposure did not affect subjective sleep quality and fatigue. Except for wake time, objective sleep measures were not related to subjective ratings. The findings underline the importance of objective sleep measurements for the examination of transportation noise effects in children.
	The effect of road traffic noise spectrum on sleep Valtteri Hongisto, Saana Myllyntausta [Abstract] Two facades having nominally equal sound insulation declaration may have very different frequency behaviors. Our purpose was to determine, how the spectrum of RTN affects sleep when the A-weighted level is constant. Twenty one participants slept three nights in a sleep laboratory in three different sound conditions: L (high-frequency prominence, 37 dB), H (low-frequency prominence, 37 dB), and Q (control, quiet, 19 dB). Conditions L and H were created by modifying a 8-hour recording of RTN using two different frequency filters corresponding to the sound insulation spectra of two constructions. Sleep quality was measured both objectively (polysomnography) and subjectively (morning and evening questionnaires). According to the subjective measures, sleep quality was worse in conditions L and H than in Q but L and H did not differ from each other. SWS duration was longer in conditions L and H than in Q. Polysomnography revealed no differences between the conditions regarding sleep duration, sleep latency, WASO, or REM, N1, or N2 durations. However, retrospective questionnaire revealed that 81% of participants assessed condition H to be the most disturbing for sleep. High-frequency RTN may be more adverse than low-frequency RTN. More research is needed on the effects of spectrum on sleep.
	Traffic Noise and its Impact on Sleep Depth Measured by the Odds Ratio Product Michael Smith, Magdy Younes, Daniel Aeschbach, Uwe Müller, Mathias Basner [Abstract] Traffic noise can lead to cortical and autonomic activation, disrupt sleep and impair physical and mental restoration. We used the odds ratio product (ORP), a validated continuous measure of sleep depth and quality based on automatic analysis of physiologic sleep data in 3s epochs, to investigate temporal changes of sleep in response to nocturnal noise events. Seventy two healthy participants slept for 11 nights in a laboratory, during which we measured sleep with polysomnography. In 8 nights, participants were exposed to 40, 80 or 120 road, rail and/or aircraft noise events at 45-65 dB LAS,max. Event-related maximum change of ORP in a 90s window relative to pre-event baseline was analysed with linear mixed models. ORP increased monotonically with LAS,max for all traffic modes, reflecting reduced sleep depth and quality, and there was a greater response to road and rail noise than to aircraft noise. There were no statistically significant effects of noise duration, sex or sleep spindle rate of occurrence on event- related ORP change. Average whole-night ORP increased with age, but was not affected by nighttime noise level (LAEq) or number of noise events. The clinical relevance of event-related elevations of ORP is currently unknown, and warrants further investigation.
	Do characteristics of short term transportation noise exposure fluctuation better predict self-reported sleep disturbances than Leq-based average noise metrics? Mark Brink [Abstract] We report on a study that aimed at establishing exposure-response relationships reflecting the per-centage highly sleep disturbed (%HSD) as functions of road traffic, railway, and aircraft noise at night time, measured by various night time noise exposure indicators at both the loudest and faintest fa-cade point of dwellings. We therefore conducted a mixed-mode representative population survey in a stratified random sample of 5592 residents exposed to transportation noise all over Switzerland, to-gether with high temporal resolution noise exposure calculations. We hypothesized that in addition to average exposure metrics (e.g. Lnight), the effects of noise on self-reported sleep disturbance can be better explained when also considering the intensity of short-term variations of noise level over time, parametrized in the Intermittency Ratio (IR) metric, which expresses the ”eventfulness” of noise expo-sure situations as the energetic contribution of individual noise events relative to the total sound en-ergy in a given measurement period. The study also explored the role of a range of coping strategies taken against noise exposure and the potential benefit of the availability of a quiet side of the dwell-ing as well as the effect of different sleep-timing patterns on exposure-response relationships.

Organization:

M. Klatte, Technische Universität Kaiserslautern C. Clark, Queen Mary University of London

	Comparing intentional switching of auditory selective attention in children and adults in an experiment suited for children Karin Loh, Edina Fintor, Sophie Nolden, Janina Fels [Abstract] A remarkable part of children’s development and educations happens in educational institutions. Acoustic environments in these institutions are usually highly complex and noisy, hence it is demanding to identify relevant target speakers and to ignore irrelevant sounds. Previous research has thoroughly analyzed auditory selective attention in adults, both in a dichotic and a binaural listening environment. Until now, there is little knowledge of auditory selective attention in children. In the present work, the original paradigm was adapted for children by using a task suited for children which included child-oriented elements. Further, the subject’s anthropometric sizes were considered for an aurally-accurate reproduction of the acoustic scene. An experiment with 24 adults and 24 children was conducted analyzing noise and noise-free conditions and various target-distractor distributions in the room among others. The result of this experiment revealed significant differences between adults and children, especially in the way auditory attention was influenced by noise.
	Effects of irrelevant background speech on verbal working memory tasks Maria Klatte, Kirstin Bergström, Thomas Lachmann [Abstract] Short-term memory for visually presented verbal items is impaired by task-irrelevant background speech that the participants are instructed to ignore. This so-called ”irrelevant speech effect” (ISE) has been attributed to interference with phonological representations, interference with representations of serial order, or attentional capture. In order to learn more about the task characteristics necessary for ISE evocation, we designed 3 verbal working memory tasks that were comparable with respect to task structure, but differed in the cognitive processes involved. In each task, sequences of 7 German nouns were presented pictorially. In Exp. 1, participants had to reconstruct the serial order of the items. In Exp. 2, participants had to decide which of the 7 words represented by the pictures had no ”partner” with the same initial phoneme. In Exp. 3, participants had to decide which of the 7 items had no ”partner” with respect to semantic category. Serial order reconstruction and phonological categorization were significantly impaired by irrelevant speech, whereas semantic categorization was unaffected. The results indicate that serial order retention is not a necessary precondition for ISE evocation. Implications for theoretical accounts of the ISE and practical implications are discussed.
	Remembering landmarks in a virtual maze: Does the disturbance impact of background speech depend on the spatial information inherent in the speech signal? Sabine Schlittmeier, Edina Fintor [Abstract] Whether a particular background sound impairs cognitive task performance depends on both sound and task characteristics. Although this so-called interference-by-process principle (e.g. Marsh et al., 2009) is not questioned in principle, it has not yet been fully empirically tested. In recent years, research on cognitive noise effects has focused on the detrimental impact of background speech’s intelligibility, semantics and temporal-spectral variability on performance in verbal tasks (e.g. verbal short-term memory, reading, writing). For visual-spatial tasks, however, the spatial characteristics of background noise should be decisive for inducing a performance decrement. We present a series of experiments in which short-term memory for visual-spatial information was explored during background speech conditions in which talkersë locations either varied or not. Considering wayfinding and orientation as everyday visual-spatial tasks, we set up virtual mazes including landmarks (e.g. a bakery, a park). Participants had to find their way through these virtual mazes and recall afterwards the spatial position of the landmarks. The effects of the different background speech conditions on visual-spatial memory will be discussed in terms of cognitive psychological implications as well as applied contexts.

Organization:

E. Van Kempen, Netherlands National Institute for Public Health and the Environment N. Riedel, University of Bremen

Aircraft noise exposure and saliva cortisol in a pooled-analysis from seven European countries Clémence Baudin, Marie Lefevre, Jenny Selander, Wolfgang Babisch, Ennio Cadum, Marie-Christine Carlier, Patricia Champelovier, Konstantina Dimakopoulou, Danny Huithuijs, Jacques Lambert, Bernard Laumon, Goran Pershagen, Töres Theorell, Venetia Velonaki, Anna Hansell, Anne-Sophie Evrard [Abstract] Introduction HYENA (HYpertension and Exposure to Noise near Airports), a well-known study on the health effects of aircraft noise, involved 4,861 participants from six European countries (except France). DEBATS (Discussion on the health effects of aircraft noise), a study with a similar protocol, included 1,244 participants from France. Saliva samples were obtained for 439 and 954 participants in HYENA and DEBATS respectively, to determine cortisol concentration as a possible marker of noise-induced stress. Associations were found between aircraft noise and cortisol levels in each study separately, but they were not consistent between the two studies. The objective of the present study was to combine datasets to improve statistical power. Methods We investigated the associations between aircraft noise levels and, the log-cortisol levels in the morning and in the evening, and the log-relative variation per hour, adjusted for relevant confounders. Results We found evidence in women for an increase in the log-evening cortisol level β=0.09 (95%CI 0.01;0.17), and for flattening of the log-relative variation per hour β=-0.09 (95%CI -0.16;-0.02), with a LDEN 10-dB(A) increase, but no significant effects in men. Conclusions These findings provide some support for psychological stress induced by aircraft noise exposure, resulting in disruption of hormonal rhythms.

Association Between Transportation Noise and Cardio-metabolic Diseases: an Update of the WHO Meta-analysis Danielle Vienneau, Ikenna Eze, Nicole Probst-Hensch, Martin Röösli [Abstract] For the development of the WHO Noise Guidelines for the European Region, meta-analyses for various cardiometabolic outcomes were conducted to derive exposure-response associations for aircraft, railway and road traffic noise. Papers published until August 2015 were considered. Since then, several new studies have been published. In the framework of revision of regulatory noise limits in Switzerland, up to date evidence for the health effects of noise is needed. Thus, the aim is to update the WHO meta-analysis for ischemic heart disease and diabetes with studies published until February 2019. The same protocol is applied as done by the WHO Environmental Noise Guideline group. Systematic literature search follows the search strategy of the WHO. Risk of bias for each study is evaluated for various design aspects. The quality of evidence is assessed using GRADE. Pooled exposure-response associations are calculated for road, railway and aircraft noise based on fixed or random-effects model depending on the outcome of the Cochran’s Q-test and the I2-statistic, which reflects the percentage of between-study heterogeneity. We will present the updated evidence and exposure-response analysis for Lden and discuss whether and how evidence for ischemic heart disease and diabetes has changed in the last few years.

Organization:

I. Van Kamp, Netherlands National Institute for Public Health and the Environment K. Kawai, Kumamoto University

	Noise and health in children with autism spectrum disorder Hidetoshi Takahashi [Abstract] Autism Spectrum Disorder (ASD) is associated with persistent impairments in adaptive functioning across multiple domains including social, communicative, occupational, and other important areas of daily life. Sensory abnormalities have often been reported as symptoms of ASD, and auditory over-responsiveness (AOR) is the most common sensory-perceptual abnormality in individuals with ASD. Recently, we reported that a greater acoustic startle reflex magnitude in response to weak stimuli of 65-85 dB was found in children with ASD compared to those with typical development, and that this index was related to various clinical features, including sensory problems, autistic traits, emotional/behavioral difficulties, and adaptive/maladaptive behaviors in ASD children. This intensity level of acoustic stimuli is frequently experienced in everyday situations including the classroom environment. Noise affects several aspects of both physical and mental health and those with AOR might be affected more by their daily environment. This would apply not only to children but also to adults working in schools, who suffer from AOR. Thus, for the healthy development of ASD children with AOR it is important to create an auditory-friendly environment including in the classroom.
	Effects of aircraft noise and living environment on children´s well-being and health Jan Spilski, Martin Rumberg, Martin Berchtold, Kirstin Bergström, Ulrich Möhler, Thomas Lachmann, Maria Klatte [Abstract] There are few and inconsistent results suggesting that chronic exposure to aircraft noise is associated with poorer quality of life or health in children. In their WHO review, Clark & Paunovic (2018) pointed out, that the current evidence is insufficient. Previous publications mainly analyzed the isolated relationships between single stressors (e. g. aircraft noise, pollution) and health in children. A combined consideration of environmental stressors and residential environment factors, such as household crowding, access to green/open space urbanity, and other urban planning data did not occur. In the NORAH study (Klatte et al., 2016), health, noise exposure, and a variety of potential moderating variables (residential environments factors) were assessed from second-graders who live in the vicinity of the Frankfurt/Main airport, Germany. We present a theoretically motivated secondary analysis of the NORAH data set. Post hoc, we linked this data with spatial and urban planning data, to model the impact of aircraft noise and children’s real life circumstances. We calculated moderated and mediated regression models and found empirical evidence for moderation effects in the relationship between noise, residential environment and well-being or health.
	Restorative soundscapes and stress recovery in Children Shan Shu, Hui Ma [Abstract] Emerging studies have shown evidence-based effects of restorative soundscapes on adults’ health. Little is known, though, about children’s physiological and emotional responses to restorative soundscapes. Therefore, the present study examined the restorative effects of soundscapes on children’s health in a simulated urban park setting. In the experiment, 53 children aged 8-12 were exposed to different soundscapes after a stressful arithmetic task. Physiological responses and emotional evaluations were measured respectively during the soundscape exposure. The results showed that all the soundscapes could decrease children’s skin conductance level (SCL) at the beginning of the recovery period, but showed significant divergent influence afterwards. In addition, different soundscapes showed marginally different effects on children’s heart rate variability (HRV) which is indicative of parasympathetic activation. Moreover, children gave soundscapes significantly different emotional ratings on pleasure and arousal, but not on dominance. Those results suggest that soundscapes such as music and stream sounds do facilitate physiological and emotional recovery on children to some extent after a psychological stressor. However, more evidence was needed to be proven in future studies.
	Associations of traffic noise and air pollution with birth outcomes in Alpine areas: Results from the UIT and BBT surveys Angel Mario Dzhambov, Iana Markevych, Peter Lercher [Abstract] The present study investigated associations between traffic noise and air pollution and birth outcomes in several Alpine areas. We used data from two representative cross-sectional studies (UIT, n = 573 and BBT, n = 518) in the Tyrol Region (Austria and Italy). Only mothers who had lived in their current residence during the whole pregnancy were included. They completed a questionnaire, and medical records were used to draw data on birth weight, low birth weight (LBW), preterm birth, and small for gestational age (SGA). Road/railway traffic noise (Ldn, Lden) and air pollution (NO2) were determined using a combination of sophisticated modelling and field measurements. In the UIT survey, higher Ldn was associated with higher odds of LBW. No associations were observed for NO2. In the BBT survey, results were not straightforward. Lden was associated with lower birth weight when adjusted for NO2 and among preterm babies, while its association with LBW was inconsistent. NO2 was associated with lower birth weight and higher odds of LBW in preterm babies. Our preliminary findings suggest that traffic emissions might increase risk of poor birth outcomes in the Alpine region.
	Effect of Sound Absorption on Children’s Behavior in Daycare rooms: a Field Experiment with Installation of Temporary Sound Absorption. Keiji Kawai, Shota Masumoto, Ryuichiro Yamane [Abstract] We have been investigating the effect of sound absorption for children in child daycare rooms and the results demonstrated a mitigating effect of bustling atmosphere in the rooms and an improvement of word intelligibility of children. In this study, following these previous findings, we attempt to find behavioral change of children in acoustically absorbed daycare rooms by a field experiment in two daycare centers with installation of temporary sound absorption. The experiment consisted of 1) teachers’ observation, where they daily evaluated children’s calmness or mood, 2) interview to teachers, and 3) behavioral observation for the concentration of children by gaze or head direction through video-analysis. As a result, while there were little difference in the teachers’ evaluation with/without absorption, the teachers, in the interview, preferred the absorbed conditions mainly because of the calmness in the room. The recorded video was divided into 1 s segments and the ratio of the segments in which the eyes or head of a child was directed at the teacher was defined as concentration ratio, and the ratio was obviously increased with sound absorption.

Organization:

S. Stansfeld, Wolfson Institute of Preventive Medicine, Queen Mary University of London I. Van Kamp, Netherlands National Institute for Public Health and the Environment

	Study on methodology to perform an environmental noise and health assessment – a guidance document for local authorities in Europe Irene Van Kamp, Dirk Schreckenberg, Elise Van Kempen, Mathias Basner, Alan Lex Brown, Charlotte Clark, Danny Houthuijs, Oscar Breugelmans, Annemarie Van Beek, Brigit Janssen-Stelder [Abstract] The Environmental Noise Directive (END) is for local authorities the most important instrument to determine the levels of noise pollution. The Directive is aimed at the evaluation of effect of measures and can make the effects of alternative measures visible. The Directive stimulates the Member States to take action to reduce the aversive effects of environmental noise. In preparation of the update of Annex III of the Directive RIVM in collaboration with international partners has prepared a guidance document taking the new WHO guidelines as a point of departure. Not only annoyance and sleep disturbance are addressed as health effects, but also cardiovascular effects and cognitive impact in children (comprehensive reading impairment). The guidance document was prepared on request of the EU commission. This paper will describe the steps of a health impact assessment one by one and explain the accompanying decisions and conditions. Next, the actual calculation methods are further explained for two indicators: the number of healthy life years adjusted for disease, disability and death (DALY) and the number of people that experiences adverse effects of noise ((NafP). Finally, as an example, the health impact of noise in Düsseldorf is presented.
	Case study of an environmental noise and health assessment in the City of Düsseldorf, Germany Dirk Schreckenberg, Irene Van Kamp, Sarah Leona Benz, Stephan Grossarth, Elise Van Kempen, Mathias Basner, Alan Lex Brown, Charlotte Clark, Danny Houthuijs, Oscar Breugelmans, Annemarie Van Beek, Brigit Janssen-Stelder [Abstract] In their recent Environmental Noise Guidelines (ENG) the World Health Organization (WHO) identified several critical health outcomes of environmental noise, among them cardiovascular diseases, annoyance, effects on sleep, cognitive impairment, and hearing impairment/ tinnitus. Aiming at protecting the population from adverse noise effects, the Environmental Noise Directive 2002/49/EC (END) is the main instrument guiding the noise mapping and the development of noise action plans in EU Member States. END Annex III describes methods of the assessment of health impacts (HIA) of noise in terms of the number of highly annoyed and sleep disturbed persons in a given area. A new study commissioned by the European Commission describes a HIA methodology for environmental noise which considers the critical health outcomes as identified in the WHO ENG. In summary, two methods, the assessment of number of people affected by different health outcomes and the environmental burden due to environmental noise in terms of the calculation of disability-adjusted live years (DALY) are described. We present a case study where both HIA methods were applied to the City of Duesseldorf, Germany, in order to describe the status quo of the health impact as well as the health impact of simulated transportation noise interventions.
	Environmental Noise Guidelines Implementation in Poland Anna Preis [Abstract] In the European Noise Directive 2002/49/EC (END) in the article 6(3) there is a statement that ”harmful effects may be assessed by means of dose-effect relations referred to in ANNEX III”. The dose-effect relations could be used to assess the effect of noise on populations. In particular, it is possible for a given value of noise index, e.g. Lden of a given noise source, to predict the percentage of people who will be annoyed or highly annoyed by this noise. The WHO environmental noise guidelines document presents the recommended Lden values for five different noise sources, which when applied could prevent people from an adverse health effects. The recommendations presented in the WHO guidelines are based on the dose-effect relations for annoyance and sleep disturbance. It seems natural that these WHO proposals should be implemented in the revised version of ANNNEX III of the END. Right now the Noise Committee is working on the new version of ANNEX III and to my knowledge representatives of EU countries vary in their views. In my talk I will present how WHO guidelines are received in Poland.
	A critical review of the basis for WHO's new recommendations for limiting annoyance due to environmental noise Truls Gjestland [Abstract] The new WHO Environmental Noise Guidelines for the European Region include recommendations for limiting noise exposure to levels without adverse health effects. This paper reviews WHO’s evidence with respect to aircraft and road traffic noise annoyance. It demonstrates that WHO’s new guidelines are based on a non-representative selection of existing studies that comprise data that cannot be generalized to the residential population at large. The new WHO recommendations are unwarranted and unsupported by the reported evidence.
	Valuing impacts of noise on health - exposure response relationships in current UK guidance and the WHO Environmental Noise Guidelines 2018 Benjamin Fenech, Georgia Rodgers [Abstract] Following the publication of the Environmental Noise Guidelines for the European Region (ENG), a lot of debate has focused on the Guideline Recommendations, in particular the specific guideline levels and the strength of the recommendations. Nevertheless it is widely acknowledged that a sizeable proportion of the population is currently exposed to noise levels above the WHO recommendations, and it is important for policy and decision makers to be able to quantify the associated burden of disease as accurately as possible. From this perspective, the systematic reviews of the scientific evidence commissioned by the WHO are an equally important aspect of the ENG. In the UK the current recommended methodology for valuing health impacts associated with environmental noise is outlined in two documents published by the Department for Environment Food and Rural Affairs (Defra) and the Interdepartmental Group on Costs and Benefits Noise Subject Group (IGCBN). [1,2] This paper will look into the studies that informed the exposure response relationships recommended by these two documents. A comparison is then made with the more recent studies that informed the ENG, followed by a discussion of which aspects of a study would make it more applicable to a local population.
	Reviewing the Current Guidance in England for the Valuation of Noise Impacts in light of the Publication of the WHO Environmental Noise Guidelines 2018 and other Relevant Evidence Hilary Notley, Alisha Iyer, Emma Powell [Abstract] In England, the current guidance for valuing impacts associated with environmental noise is published by the Department for Environment Food and Rural Affairs (Defra) and the Interdepartmental Group on Costs and Benefits Noise Subject Group, (IGCB(N)) [1]. The development of the World Health Organisation (WHO) Environmental Noise Guidelines [2] followed a rigorous methodology and their recommendations are based on systematic reviews published separately [3]. The systematic reviews considered evidence published between 2000 and 2014 - 2016 (health outcome dependent). There have been several high quality studies published since those considered by the WHO that may further develop the understanding of the link between noise exposure and health. This paper describes the steps that the IGCB(N) are taking to determine whether they should update their current recommendations. The paper describes the commissioned reviews of evidence relating to health effects and efforts to assess a number of potential effects that are outside of the scope of the WHO, such as the benefits of quiet areas/tranquillity and the impact on productivity. The paper further describes how the valuation of any effects will be considered. [1] https://www.gov.uk/guidance/noise-pollution-economic-analysis [2] WHO Regional Office for Europe Environmental Noise Guidelines for the European Region http://www.euro.who.int/__data/assets/pdf_file/0008/383921/noise-guidelines-eng.pdf?ua=1 [3] https://www.mdpi.com/journal/ijerph/special_issues/WHO_reviews

Organization:

T. Masayuki, Kyushu University M.E. Altinsoy, Technische Universität Dresden, Lehrstuhl für Akustik und Haptik

	Analysis of vehicle horn use and factors at intersections in an urban area of Taiwan Takada Masayuki, Shoki Tsunekawa, Kazuma Hashimoto, Tamaki Inada, Yoshinao Oeda, Katsuya Yamauchi, Ki-Hong Kim, Shin-Ichiro Iwamiya [Abstract] In the urban areas of many countries, noise pollution is generated by frequent use of car and motorbike horns. The final goal of the present study is to clarify the causes of horn use and to propose a transportation system to suppress horn use. To this end, we have been investigating relationships between horn use and factors such as driver awareness and behavior and vehicular environment. Our previous study investigated causal relationships between horn use and the vehicular environment in Taiwan and found that horn use was frequent when gridlock formed at intersections. The present study analyzed details of gridlock situations during which vehicle horns were used at three intersections. We found many cases where honking began after vehicles entered the intersection to turn left during the green-light phase. In particular, drivers often used their horns when they had to wait for more than 4 s for the vehicles in front of them to start moving at the beginning of the green-light phase for turning left. Logistic regression analysis revealed that the duration between the signal turning green and the vehicle in front of them starting to move significantly affected the probability that a driver would use a vehicle horn.
	Electric and Autonomous Vehicle: from Sound Quality to Innovative Sound Design Nicolas Misdariis, Andrea Cera, William Rodriguez [Abstract] This study follows long-term lines of research and realizations done within the frame of electric vehicle sound design. It currently deals with the most up-to-date issue addressing the near future mobility: Autonomous Driving (AD). It is implemented in a 2-year collaboration with a french car manufacturer that deployed a high-tech demo car where the sonic dimension was explored in a spatial, and a multisensory, approach. Specific hardware devices (multi-channel diffusion system) and software protocols (spatialization and vehicle data transmission) made it possible to study and realize numerous and accurate internal / external warning sounds and auditory Human- Machine Interfaces - and finally to contribute to a kind of ”passive branding” of the artifact. The Autonomous Driving condition was especially considered as it corresponds to an unprecedented user experience in terms of human-centered process, where sounds could particularly play a role for informing and, if necessary, reassuring the driver and passengers. The paper presents a thorough analysis of the project: initial requirements and data, multi-disciplinary framework, technical details, sound concepts and realization. It also opens onto reflexive and prospective considerations about the global notion of sound quality that especially look at possible forms of ”low-intrusiveness” in sound design, and their related properties.
	Design of Acoustic Vehicle Alerting System Sound Assuming Listening Situation of Pedestrians Hiroshi Matsuda, Masato Suzuki, Nobuo Machida [Abstract] The role of acoustic vehicle alerting system (AVAS) sound is to tell pedestrians the approaching of quiet vehicles such as electric vehicles or hybrid-electric vehicles (HEV). We propose the AVAS sound that satisfies both of comfortability and perception. The comfortability of AVAS sound in this study means pedestrians do not feel noisy. The perception of AVAS sound means pedestrians to notice of the approaching quiet vehicles at safely avoidable distance. The AVAS sound model produced in this study has two peak frequencies. These models are two kinds of the steady sound with constant sound level and the fluctuating sound formed by modulating amplitude of steady sound. We measured impression of AVAS sound and perceptual distance. The impression of AVAS sound was measured by the psychological questionnaire using SD method. The perceptual distance: the distance between subject and HEV when the subject notices the approaching of HEV emitting the AVAS sound was measured in a test road. We also measured the perceptual distance of different listening situation such as standing or smartphone operation during walking. We show the sound characteristic to improve the comfortability and the perception of AVAS sound, and the effect of the listening situation on the perceptual distance.
	Relationship between acoustic characteristics and impression for warning sounds on electric vehicles Nozomiko Yasui [Abstract] The motor sound on electric vehicle is quiet at low speeds. Thus, pedestrians have difficulty detecting those vehicles approaching them. Although those vehicles were designed to play an alert sound to solve this problem, it has not been solved yet. When the sound is designed, it should be concerned not only detectability of approaching quiet vehicles but also impression of the sound. For pedestrians, it’s important to make it easier to recognize quiet vehicle. Also, warning sounds shouldn’t contribute to traffic noise annoyance. Our previous studies found that acoustic characteristics of amplitude fluctuation are effective to make them detect approaching vehicles. Here, this study evaluates impressions of those fluctuated sound, investigates relationship between the impression and those characteristics. The impressions of synthesized warning sounds that have periodic and non-periodic amplitude fluctuations were measured by semantic differential method. The obtained data were analyzed by factor analysis. The results revealed that characteristic of amplitude fluctuation influences the factors of ”familiarity”.
	Pedestrian awareness of the approach of quiet vehicles: Effect of approach informing sound and designing awareness Takahiro Kosuge, Tomoya Tsujinaga, Toki Kobayashi, Mariko Tsuruta-Hamamura, Hiroshi Hasegawa [Abstract] Hybrid and electric vehicles are quieter than combustion- engine vehicles. To prevent accidents resulting from this quietness, an acoustic vehicle alerting system generates an approach informing sound (AIS). However, the recognition rate of the AIS is not high (∼40%) and it might be difficult for pedestrians and others to notice the approach of quiet vehicles via hearing alone. To improve AIS recognition, we designed an AIS based on onomatopoeic sounds. A subjective evaluation experiment using these sounds revealed that people tended to evaluate noisy and rough sounds, such as those similar to the sound of a car traveling or an engine, as a good indication of a car traveling. Another experiment investigated the sounds that pedestrians actually use to recognize the approach of quiet vehicles. It was found most participants recognized the approach of a quiet vehicle not by the AIS but by the sounds emitted from the vehicle, such as road and wind noise. This tendency was also observed after participants correctly recognized the AIS. However, some participants reported that their awareness of AIS changed and they became more aware of AIS in daily life after the experiment. Designing awareness is effective and necessary for the design of AIS.
	Effects of localization control of warning sound combined with visual information in vehicle cockpit Kiichi Naka, Katsuya Yamauchi, Nobuaki Tanoue, Ayumu Kawata [Abstract] Recently, many kinds of information are provided to drivers because of the improvement of ADAS (Advanced Driver Assistance System). Amount of information in vehicle cockpit seems to be increased due to the development of automobile technology such as autonomous driving. It is very important to consider the design of appropriate auditory information to avoid overlooking visual information cues. In this paper, we examined the effect of localization control of warning sounds to determine the direction of visual information using a driving simulator. The number of overlooked visual information cues and subjective task load were significantly decreased by the localization control in comparison with the no localization and no sound conditions. A driver could recognize the direction of visual information without removing their gaze from frontal view so that a driver also could find the visual information rapidly and accurately, and could watch the frontal view more long duration. It led to steady driving performance such as steering and speed control. In contrast, these effects were diminished if the timbre of the signal was difficult to localize.

Organization:

T. Toi, Chuo University W.-H. Cho, KRISS S.-H. Shin, Kookmin university I. Jung, Hyundai Motor Company

	Evaluation of Independent Sound Zones in A Car Ji-Ho Chang, Wan-Ho Cho [Abstract] Recently sound field control techniques have been applied to vehicles that generate different sound zones in a space with loudspeaker arrays. A typical example in a car is to provide a driver and a passenger with different sound such as voice-guided navigation, music, or telephone voice. Acoustic contrast has been widely used as a performance index in sound field control that effectively shows the difference between sound levels in the acoustically bright zone and the dark zone. However, these sound levels are usually measured with microphone arrays that are placed in the sound zones without the listeners, and averaged. These values can be different from what the listeners would have at their ears. This study compares the acoustic contrasts derived using microphone arrays with those using dummy head microphones at several positions and those using binaural microphones mounted in several users’ ears. From this comparison, this study attempts to propose a simple and effective evaluation method of the independent sound zones.
	Sounds for Enhancing Energy Efficient Driving: A Simulator Pre-Study Arne Nykänen, Johan Fagerlönn, Stefan Lindberg, Roger Johnsson, Sathish Kumar Ramanathan [Abstract] Computer game technology was used for rapid prototyping of a sound based interface encouraging truck drivers to drive energy efficiently. The design process was inspired by user-centred agile methods. Interior sounds were made interactive by actively controlling them based on speed, engine speed, torque and acceleration. User feedback was collected at an early stage through contextual enquiry sessions during simulated driving. Based on the feedback the sounds were adjusted and effects on driver behaviour were measured in a simulator experiment. The effects on driver behaviour were small and not statistically significant (p>.05). However, assessments of preference and informativeness showed statistically significant differences between the design concepts (p<.05). The qualitative part of the study showed that the use of game technology for enabling assessment of interactive sounds in early design phases was useful and allowed for getting users into the loop early on. The framework was found to be suitable for designing interactive sounds, and the data collected provides insight into driver responses to using active noise control as a means for providing information to the driver.
	Investigation of Effect on the Acoustic Transfer Function in a Vehicle Cabin According to Change of Configuration Wan-Ho Cho, Ji-Ho Chang [Abstract] Recently, the various audio information are presented in a vehicle cabin not only for a driver but also other passengers and various sound field control methods had been applied for various purpose such as personal audio. Basically, the measurement of acoustic transfer function in a vehicle cabin is relatively easier than other living spaces because of its small volume and movement of objects. However the effect of configuration cannot be ignored due to the relatively large volume change, especially related to the change of number of passengers. Here, the effect on the acoustic transfer function in a vehicle cabin according to change of configuration is observed and method to estimating robust solution is investigated. The solutions estimated by using the transfer matrix measured by planar arrays, dummy heads, and arrays surrounding head position with dummy heads are compared by changing the number of passenger. To observe the robustness of estimated solution, the solutions are applied to the different conditions not used to estimate the solution and the degradations are compared for each case.
	Factors Analysis of Gear Sound by Using Numerical Simulation Chengfeng Pu, Takafumi Takeda, Toru Hama, Takeshi Toi [Abstract] As an approach to identification of a factor that contributes to gear sound, In the quasi - static loading condition of low speed operation, estimation of gear transmission error of spur gear was investigated from combination of vibration measurement and numerical simulation. Furthermore, build a numerical simulation approach to analyse the prediction and contribution of vibration and sound under high speed operation conditions.
	Design of Optimal Car Interior Sound Improving Driving Feeling Su-Ho Cha, Sung-Hwan Shin, Jae-Ho Yang [Abstract] The virtual driving sound has been used as a tool improving the driving feeling of passenger car. The purpose of this study proposes is to propose a strategy generating virtual driving sound that is closely connected with the driving condition. In this end, previous study ascertained that dynamic feeling of car interior sound was related to the change of tone color during the acceleration. The tone color was affected by the strength of the firing order components. And then, it was analyzed that the relation between the dynamic feeling obtained from the subjective listening test and the acceleration condition of passenger car and as its result, it was identified when a virtual sound should be generated to improve the driving feeling such as the starting and stopping points of virtual sound and its reproduction duration. Next, optimal levels of the firing order components are found out by subjective listening tests. Lastly, the effect of the proposed strategy generating virtual driving sound is investigated. It is expected that the method can improve driving feeling related to not only dynamic but also pleasant perception.

Organization:

S. Sakamoto, Institute of Industrial Science, The University of Tokyo K. Yamauchi, Kyushu University, Faculty of Design

	Road Traffic Noise Prediction Model “ASJ RTN-Model 2018” Proposed by The Acoustical Society of Japan – Part 1: Outline of the calculation model Shinichi Sakamoto, Yasuaki Okada, Akinori Fukushima, Toshio Matsumoto, Terutoshi Tajika [Abstract] The Acoustical Society of Japan (ASJ) has published a new version of road traffic noise prediction method ”ASJ RTN-Model 2018” in this April. The technical committee on road traffic noise in the Acoustical Society of Japan (ASJ) had been working for structuring the road traffic noise prediction models in conformity with the times for decades, and ”ASJ RTN-Model 2018” is an up-grade version of the previous model proposed in 2014. In developing the new version, existing knowledge is widely taken into account, in particular, specification of sound power levels of vehicles and calculation methods for sound propagation are improved in wide range. This new model is introduced in five papers with the same main title in this congress. In this paper, the framework of the model, the calculation principle, the general procedure of prediction calculation, the outline of the revision, are described.
	Road Traffic Noise Prediction Model “ASJ RTN-Model 2018” Proposed by The Acoustical Society of Japan – Part 2: Calculation Model of Sound Emission of Road Vehicles Yasuaki Okada, Akinori Fukushima, Katsuya Yamauchi, Shinichi Sakamoto [Abstract] The Acoustical Society of Japan (ASJ) has published a new revised version of road traffic noise prediction method ”ASJ RTN-Model 2018”, in which the calculation formula for the A-weighted sound power level of each type of road vehicle is specified. The sound power level is given simply as a function of the running speed in consideration of practicality and convenience. The Research Committee in ASJ has been accumulating new data about noise emission of vehicles running on the dense and porous asphalt pavements over the last decade. As a results, the sound power level of light vehicle (includes passenger cars, hybrid vehicles and small-sized vehicles) for the dense asphalt pavement was changed to low value. In addition, a improved calculation formula of sound emission on the drainage asphalt pavement were introduced. In this paper, the revised sound power level and spectrum models for road vehicles using the newly acquired data are described.
	Road Traffic Noise Prediction Model “ASJ RTN-Model 2018” Proposed by The Acoustical Society of Japan – Part 3: Calculation model of sound propagation Akinori Fukushima, Shinichi Sakamoto, Yosuke Yasuda, Takatoshi Yokota [Abstract] As the third part of ”ASJ RTN-Model 2018”, calculation model of sound propagation is presented. The model is an up-grade version of the previous model that was proposed in 2014. This model is basically developed as a practical calculation model based on Geometrical Acoustics. The calculation equations in this model are defined on the base of experimental or numerical analysis data. The overall values of the A-weighted sound pressure level propagated from a vehicle are directly calculated by considering the frequency characteristics of vehicle noise. This calculation model consists of shielding effects by several type barriers, excess attenuation due to ground surface, atmospheric absorption effect, sound reflection and meteorological effects. Based on newly obtained knowledge, several improvements were made on the calculation method to develop the model. The procedures of application to roads with special cases such as road tunnel, semi-underground road and double deck viaduct road are also included.
	Road Traffic Noise Prediction Model “ASJ RTN-Model 2018” Proposed by The Acoustical Society of Japan – Part 4: Accuracy Verification of a Practical Method for Areas behind Buildings in Urban Districts Ken Anai, Toshio Matsumoto, Takatoshi Yokota, Shinichi Sakamoto [Abstract] The purpose of this study is verification of ASJ RTN-Model for areas behind buildings, which is named in this paper as a practical method, by comparing predicted noise levels and measured ones in urban districts. Two kinds of verification are demonstrated in this paper. In the one kind of verification, max noise levels in unit patterns by running an inspection car in steady speed on a flat road or a cut road facing at detached residential areas are focused on. In the other kind of verification, equivalent noise levels along two city roads are predicted by the practical method with road traffic census data. As a consequence, differences between the predicted levels and the measured ones are within plus and minus 3 dB in the max noise levels case, and the differences are from plus and minus 3 dB to 5 dB in the equivalent noise levels case.
	Road Traffic Noise Prediction Model “ASJ RTN-Model 2018” Proposed by The Acoustical Society of Japan – Part 5: Study on Prediction Accuracy Katsuya Yamauchi, Terutoshi Tajika, Akinori Fukushima, Ken Anai [Abstract] This paper introduces the study on prediction accuracy of the ASJ RTN-Model 2018. The authors provide the report as a part of the series of five papers which has same main title in this congress, on behalf of the research committee on road traffic noise in the Acoustical Society of Japan. It is very important to examine the prediction accuracy of the new model, because some assumptions are included in modeling the source power level of each type of vehicles and in simplifying the calculation of noise propagation. Some examination of the correspondence between the predicted values and measured traffic noise levels are introduced in this paper. The accuracy of the method of prediction for noise behind building complex are also examined. The measured values were obtained recently in Japan on flat and strait section of general roads and on expressways. This paper also introduces the examination on assumed causes of errors that are thought to be important in considering prediction accuracy, such as the condition setting of hypothetical traffic lanes, vehicle type classification, and some typical causes of uncertainty on actual measurement.
	Experimental Modelling of Tyre/Road Noise from Road Texture Spectra on Rubberized Road Surfaces Alessandro Del Pizzo, Gonzalo De Leon, Luca Teti, Francesco Bianco, Antonino Moro, Luca Fredianelli, Gaetano Licitra [Abstract] The present work studies the correlation between road texture and tyre/road noise from an experimental point of view, by analysing road texture and CPX measurements on rubberized road surfaces. Tyre deformation has also been addressed using the indenter method proposed by Goubert and Sandberg. The correlation coefficient between texture and noise bands was calculated using data deriving from CPX measurements and enveloped texture spectra. Results showed that tyre/road noise can be separated into low, middle and high frequency components, which correlate differently with road texture bands. This frequency separation was used to develop two different linear models that relate low and high frequency noise to road texture one-third octave bands.
	An Australian case study on the estimation of heavy vehicle noise emission on grade Jeffrey Peng, Daipei Liu, Jeffrey Parnell, Nicole Kessissoglou [Abstract] Heavy vehicles are considered the primary determinant of night-time noise disturbance, particularly in rural areas or for residents along freight routes. To capture the dynamic influence of heavy vehicles associated with variation in speed on noise emission, heavy vehicle kinematic variables need to be incorporated within a road traffic noise emission model. These kinematic variables in turn assist with accurate estimation of engine noise and rolling noise. The American FHWA TNM road traffic noise model is the only existing prediction method that considers the driving speed profiles of articulated trucks. However, it can only consider a fixed set of speed profiles based on a single heavy vehicle power- to-weight ratio. As such, the current model is limited and does not accurately represent longer and heavier vehicle combinations. In this work, the FHWA TNM method is modified to include the kinematic equations for a heavy vehicle on grade. The model is applied to heavy vehicles commonly utilized in Australia. A case study based on a principal freight route in New South Wales, Australia, is presented to illustrate the predicted variation in engine noise and rolling noise throughout the heavy vehicle’s journey.
	Vehicular traffic noise in bus station users modeling and prediction, based on the analysis of direct and specular paths Dayane Cristina Lima Estercio, Paulo Fernando Soares [Abstract] Bus stop for public transportation systems are structures where users are exposed to intense noise levels and are poorly evaluated in surveys. Therefore, this research aims to develop a mathematical model of noise level prediction at bus stops, in order to contribute to the characterization and evaluation of the influence of traffic noise level in these buses. In order to achieve the purpose of the research, a prediction model was developed based on the direct and specular reflections. A standard bus stop model was adopted on a three-lane road, simulating the users who are standing and sitting The sound source was positioned on the axis of each strip, every five meters. In total, 378 source positions were evaluated. The descriptors were the sound pressure level (SPL) and the sound equivalent level (LAeq). For the validation, the Anderson-Kurze model and the t-Student test were applied. The results presented better correspondence between the developed model and the reference model in the 1000Hz range and in the 500Hz frequency there was a greater variance between the applied models. As a preliminary study, the research presented satisfactory results, which may contribute to future research.

Organization:

M. Wegerhoff, HEAD acoustics GmbH J.-G. Ih, Acoustics Laboratory, Dept. of Mechanical Engineering, KAIST

	Transfer Path Analysis of Rumbling Noise in a Passenger Car Based on Measured In-Situ Blocked Force Sang Kwon Lee, Taejin Shin, Yeonsoo Kim, Kanghyun An [Abstract] The control of rumbling noise is one of the major strategic targets of interior sound quality inside the cabin of a passenger car. To effectively control rumbling noise in a passenger car, the transfer path of the rumbling noise should be initially identified. It is known that the major source of this noise the combustion force of an engine. The combustion force excites the engine and induces vibrations of the powertrain. These vibrations are then transferred to the body of the vehicle via its structural transfer path. Moreover, the vibrations of the vehicle’s body emit internal vibra-acoustic noise. This noise is often referred to as the rumbling noise due to the structural borne path. If there are structural resonances among the structural paths such as the engine, transmission, mount bracket, suspension, and the vehicle’s body, the rumbling noise could be amplified. Recently, blocked force transfer path analysis (BF-TPA) was introduced and this approach does not require the decoupling of the substructures. In this study, we identify the structure-borne path of rumbling sound based on blocked force transfer path analysis (BF-TPA) in a passenger car. In addition to identification, the passive control method for rumbling sound is presented.
	Combining Structural Modification with In-Situ Transfer Path Analysis to Solve Noise and Vibration Problems Andrew Elliott [Abstract] In-situ TPA measurements provide blocked forces that can be used to diagnose structure borne noise problems in vehicles. One of the main advantages of in-situ TPA, also known as blocked force TPA, is that all measurements are made with the source and receiver in a coupled state making measurements easier and ensuring representative operation of the vibration source. Perhaps the greatest advantage however is that the blocked forces obtained, unlike contact forces, are an intrinsic property of the vibration source which means they remain valid for modified or completely different receiver structures. In the paper, real world practical examples that exploit the advantages of in- situ TPA in terms of structure borne noise prediction and structural modification are presented. The first example is of a complex vibration source coupled at multiple points through six degrees of freedom per interface to a test bench that is subsequently built into a new assembly for which structure borne noise predictions are made. A further example is then used to demonstrate how structural modifications can be made to the receiver side of an assembly for the purposes of troubleshooting noise and vibration problems using in-situ TPA.
	In-Situ Transfer Path Analysis of Multiple Vibration Sources in a Complex Source- Receiver Assembly Lucy Susan Barton, Andrew Elliott, Andy Moorhouse, John Smith [Abstract] Depending on where the source-receiver boundary is defined within an assembly, an airborne sound source can also be considered as a structure borne sound source. For example, if the internal forcing mechanisms within a machine are regarded as the sources, the rest of the assembly including the surrounding air can be considered the receiver. If the structure borne noise resulting from these internal forcing mechanisms can be predicted in real-time, the radiated noise can potentially be predicted and managed- either actively or passively- using numerous approaches. The primary aim of the work presented in the paper is the development of an accurate method for the real-time prediction of structure borne noise, whilst continually monitoring the reliability of these predictions. To facilitate this process, it is necessary to also monitor the stability of the vibro-acoustic frequency response functions that relate the internal forcing mechanisms to the receiver position, and to eliminate the influence of background noise resulting from noise or vibration sources other than those of interest. Presented in the paper are the results of a laboratory transfer path analysis case study of a source-receiver assembly composed of multiple sources attached to a single receiver structure.
	Analysis of a vibrating structure as an airborne sound source by means of matrix inversion Serafima Anisovich, Roland Sottek, Matthias Wegerhoff [Abstract] The decomposition of a complex airborne sound source into a number of simple sources has found wide application in transfer path analysis. One of the most advanced ways to solve this task is to use the matrix inversion method. This method works well for describing concentrated sources like monopoles. In practice, however, it is also desirable to analyze distributed sources such as a vibrating structure (e.g., the housing of an engine). A large number of grid points, representing emitting partial surfaces, and an even larger number of measurement points complicate the task. The large matrix of transfer functions between source points and measurement points that need to be inverted is often ill-conditioned. Mathematical approaches such as regularization techniques are required for the matrix inversion. Understanding the ways to apply these approaches and their physical meaning is one of the most important steps in finding suitable solutions to the mathematical inversion problem without losing physical information. In this paper, some of the most common matrix inversion strategies are analyzed for the example of a vibrating plate.
	Can I trust my TPA results? Frank Jürgens, Christian Nettelbeck, Philipp Sellerbeck [Abstract] In the automotive development process, Transfer Path Analysis (TPA) is a widely known tool for troubleshooting purposes of existing vehicles or for the prediction of vehicle interior noise based on test bench or CAE data. There are many different approaches, each having its pros and cons. Either determining interface forces while the source is disconnected or equivalent forces for in-situ condition; measured in the vehicle or at the test bench; calculated in frequency domain or in time domain. As different as the approaches are in detail, one question they all have in common: How reliable are their results? Within this paper, the need of a validation and refinement process of the TPA model - independent of the specific TPA method - is discussed and illustrated by typical application examples. New tools like the Mosaic View for the visualization of the crosstalk and useful procedures like the Root Cause Analysis and the Model Robustness Check are exemplified. Furthermore, these technologies are evaluated to show their potential for the identification of errors in the TPA model and how to deal with them in order to increase the liability.
	Physics-Informed Transfer Path Analysis with Parameter Estimation using Gaussian Processes Christopher Albert [Abstract] We apply a variant of Gaussian process regression to augment experimental data from transfer-path analysis (TPA) by known information about the underlying physical properties of the system under investigation. The approach can be used as an alternative to model updating and is also applicable if no detailed simulation model of the system exists. For vibro-acoustic systems at least three features are know. Firstly, observable quantities fulfill a wave equation or a Helmholtz-like equation in the frequency domain. Secondly, the relation between pressure/stress and displacement/velocity/acceleration are known via constitutive relations involving mass density and elastic constants of the material. The latter also determine the propagation speed of waves. Thirdly, the geometry of the system is often known up to a certain accuracy. Here we demonstrate that taking into account this information can enhance TPA results and quantify their uncertainties at the same time. In particular this is the case for noisy measurement data and if material and geometry parameters are (partly) unknown. Due to the probabilistic nature of the procedure unknown parameters can be estimated, making the method also applicable to material characterization as an inverse problem.
	Obtaining method of high contributing body and frame vibration behavior to road noise using principal component contribution analysis Hiroki Taguchi, Kanon Nukata, Junji Yoshida [Abstract] Recently, principal component (PC) contribution analysis utilizing operational TPA was developed to find out high contributing vibration behavior of the target structure to the interior noise. However, in case a vehicle is composed of body and frame, identifying the high contributing part between them is difficult using this method. In this study, we then attempted to propose an identification method of the high contributing part between body and frame using the PC contribution analysis sequentially. In the test, a simple vehicle model was employed and vibration at multiple points on the body and frame were measured simultaneously with the interior noise when random input signals was given under the tire. Subsequently, PC contribution analysis was applied in two steps. In the first step, high contributing body vibration behavior to the interior noise was extracted and then, high contributing frame vibration behavior to the important body mode was obtained in the second step. As the result, the frame vibration behavior was found to be the main part increasing the interior noise at the target frequency. Finally, the interior noise at the target frequency could be decreased by applying intensive countermeasure to the high contributing frame vibration behavior.

Organization:

S. Lutzenberger, Müller-BBM A. Kuijpers, M+P

	The Application of Dither for Suppressing Curve Squeal Wolfgang Kropp, Arthur Aglat, Jannik Theyssen, Astrid Pieringer [Abstract] Curve squeal is a highly disturbing tonal sound generated by vehicles like railways, metros or trams, when negotiating a sharp curve. The probability that squeal occurs increases with reduced curve radius of the track. Curve squeal noise is attributed to self-excited vibrations caused by stick/slip behaviour due to lateral creepage of the wheel tyre on the top of the rail. With respect to the enormous number of the rolling stock units and the long lifetime of waggons there is an urgent need for a cheap and simple retrofitting measure to reduce curve squeal. The main objective of paper pro is therefore to investigate the potential to reduce curve squeal by means of active control in the form of dither in an efficient and robust way. Dither control has been applied in the field of mechanical engineering for systems including non-linear components. There it has been shown to suppress self-excited oscillations very efficient. The control is an open loop control. It consists in adding a forced vibration to the vibrational system. The demand on this additional signal is that it is higher in frequency than the friction-induced response. From a physical point of view, dither control modifies the effective friction characteristic.
	The Effects of Weather Conditions and Wheel Wear on Curve Squeal Thomas Maly, Florian Biebl, Michael Ostermann [Abstract] The mechanisms of curve squeal generation (lateral stick-slip effect on the top of the rail and/or friction of the wheel flange) are well known. Even if various influencing parameters were already investigated mostly on a theoretical level, in real train operation the parameters are in general not measurable in their entirety and their effects due to the complex interdependencies between them are often not predictable. That is why curve squeal generation often seems to have a random character. As a continuation of previous studies the aim of the work presented was to investigate the influence of weather condition changes over a whole year and wheel profile parameters changes due to wear on the occurrence frequency of curve squeal. For this, measurements of the pass-by noise emissions and the weather conditions in a narrow curve (radius 230m) of a Vienna suburban train line were carried out over a period of 11 months. Due to a previously developed algorithm for automatic detection of curve squealing more than 20000 data sets were available for subsequent statistical analysis. In the paper the dependencies of squeal frequency on rain, air humidity, dew point, and rail temperature, as well as wheel parameters are presented in detail.
	Another Halving of Rail Freight Traffic Noise by Wheel Absorber Martin Fehndrich [Abstract] By 2020 the freight traffic noise in Germany and Switzerland should be halved by changing cast iron brake blocks wheels to composite brake blocks. By that levels 93 dB(A) should become 83 dB(A), which is the limit defined in the TSI Noise. A further halving reduction would lead to 73 dB(A), which is part of a SBB Cargo vision for its future fleet. The fact this reduction is possible is shown by first results of the SBB Cargo’s 5L project, where with the help of a number of suppliers a train with 16 freight wagons with a series of different noise-reducing measures (e.g. disc brakes instead of block brakes) was built, which is tested now in regular traffic. Since the wheel is still the predominant source of sound emission in railway traffic, the biggest impact is reducing the wheel noise. In SBB Cargo’s 5L project wagons with BVV wheels and BVV wheel absorbers sound levels of 73 to 75 dB(A) and an equivalent of no more than 75 dB TSI level could be achieved.
	Pass-By Noise Source Identification for Railroad Cars using Array Measurements Hans Rudolf Graf, Christian Czolbe [Abstract] Microphone arrays are suitable for acoustic source localization. For good accuracy, adequate time averaging is required. With fast moving sources, the time interval during which a source remains within a focal spot is very short, resulting in poor accuracy. By tracking and averaging the sources during pass- by, available averaging time and consequently accuracy can be substantially improved. The technique is demonstrated by pass-by measurements of innovative freight cars investigated as part of the research project www.innovativer-gueterwagen.de by the Federal Ministry of Transport and Digital Infrastructure (BMVI). The measurements were carried out in Spring 2018 on the test track in Wildenrath, Germany. In the post-processing, the acoustic source intensity maps were integrated for areas of interest, such as the wheelsets, rail sections and frameworks, resulting in sound power level of the corresponding components. Results of selected noise reduction measures will be presented. Spatial resolution of arrays is rather limited at low frequencies. Therefore, noise sources of the rail and wheel cannot be clearly separated based only on array results. Additional signals of structure borne noise of the rail may be used to improve the separation of wheel and rail related sound radiation.
	Rail roughness surveys – a tool for effective environmental noise control Lisette Mortensen, Stig Junge [Abstract] A mobile method developed by Sweco for estimation of rail roughness level on a large scale is presented. The method is suitable for countrywide surveys and the data has proved to be very useful in the planning of track maintenance both in terms of environmental noise and rail quality. During the last two years roughness measurements have been carried out twice on the Danish railway network and all roughness data from these campaigns have been stored in Banedanmark’s track measurement database. Banedanmark has developed a tool to rank track sections according to rail quality and population density in residential areas to most effectively reduce the amount of environmental noise from the track and to document the effect of noise reduction measures. The measurement method uses measurement of noise from the contact point between wheel and rail to derive an estimate of the rail roughness level and is based on the transfer function from rail roughness to acoustical noise. The method has been tested at measurement train speeds from 60 - 120 km/h and shows a high degree of reproducibility. Banedanmark has given these measurements a significant role in planning the large-scale maintenance operations of grinding and milling.

Organization:

C. Spehr, DLR T. Geyer, BTU Cottbus-Senftenberg

	Aircraft noise: Conversion of an existing to a desired number of subtracks with identical lateral dispersion to obtain smooth noise contours Olivier Schwab [Abstract] In aircraft noise calculations, the dispersion of flight tracks on a certain air route is often represented by a backbone track and a number of laterally displaced subtracks, where each track is assigned a fraction of the route’s movements to obtain a certain lateral distribution. For routes with a large dispersion in the flight tracks, the number of subtracks should be larger than for routes with small dispersion to obtain smooth footprints (i.e., single event level grids) in both cases. However, if subtracks are drawn by hand using expert knowledge, e.g. because no reference radar data is available, the number of available tracks is usually limited. Here, it can be useful to be flexible in the creation of additional and/or modification (smoothing) of existing subtracks. In this contribution, we present a method to construct a different (usually larger) number of subtracks based on a set of existing subtracks by geometric matching of the subtracks with the backbone track. The construction assumes identical lateral dispersion of the converted tracks as the original subtracks. We discuss example cases featuring curves with tight turns and non-symmetric subtracks with respect to the backbone track, and compare resulting footprints with those of the original tracks.
	SonicBAT: Some Highlights and Subsequent Developments Victor W. Sparrow, Trevor Stout, Kevin Bradley, Christopher Hobbs [Abstract] The Sonic Booms in Atmospheric Turbulence (SonicBAT) project concluded in 2018. The overall goal of SonicBAT was to develop tools to predict the influence of atmospheric turbulence on shaped sonic booms for newly proposed low-boom supersonic passenger aircraft. There was a substantial experimental component to obtain statistically-useful datasets of recorded sonic booms in both dry and humid environments. The project was carried out by a large team, and the authors of this paper are a small subset of the individuals involved. The purpose of this presentation is to highlight some of the main findings of SonicBAT and to provide pointers to the primary technical report and other publications that are now coming available, both from SonicBAT and from subsequent findings. Overall, SonicBAT was successful in developing multiple ways of predicting the influence of atmospheric turbulence on the propagation of arbitrarily shaped sonic boom waveforms as they propagate through the earth’s planetary boundary layer. The influence of turbulence is profound for N-wave sonic booms but is reduced for low-boom sonic boom signatures. [Work supported by NASA.]
	Calculation of the Fan Rotational Speed Based on Flyover Recordings for Improving Aircraft Noise Prediction Models Roberto Merino-Martinez, Mirjam Snellen, Dick G. Simons [Abstract] Noise control regulations in the vicinity of airports usually rely on the estimations of aircraft noise levels by noise prediction models. Current best-practice noise contour calculation methods assume default engine thrust values depending on the engine type and the altitude of the aircraft. These prediction tools provide a single noise level for a certain aircraft in a certain flight phase and at a specific distance from the observer. In practice, however, changes in the thrust occur and cause variations in the noise levels of several decibels. In this paper, an approach is presented to estimate the fan rotational speed N1% (and hence the thrust) directly from flyover audio recordings. This method estimates the blade passing frequency (BPF) of the fan by searching its characteristic tonal peak (and its higher harmonics) and accounting for the Doppler effect. This method was applied to more than a thousand measurements of Airbus A330-300 and Boeing 777-200 aircraft. The results show a significant correlation between the recorded noise levels and the fan rotational speed, explaining up to 45% of the variability in noise levels. Considering the calculated N1% values in the noise prediction models notably increases the agreement with the recorded noise levels.
	Numerical and Experimental Investigation of the Turbofan First Booster Stage Tone Noise Victor Mileshin, Sergey Pankov, Anton Rossikhin [Abstract] An important problem of turbofan design is maintenance of low noise emission. One of the components of turbofan noise is the tone noise arising due to inter-row interactions in the booster stages. For modern turbofans this component of noise can be significant at approach certification point. The investigations of the tone noise of the first booster stage have been carried out in CIAM during last several years. They were performed using frequency domain method for simulation of tone noise of multistage turbomachines developed in CIAM. The advantage of the method is the high speed of calculations; however it relies on some ad hoc assumptions about the process of interactions between flow field disturbances and blade rows. This work presents the results of the tone noise calculation of the first booster stage using time domain approach based on multirate Runge-Kutta scheme. The result, obtained in the calculation, is independent from any assumptions about the interactions in the multistage turbomachine. This allows verifying the results of previous calculations. In general satisfactory correspondence were found between the results of computations in time and in frequency domains and between them and experimental data obtained in the CIAM C-3A acoustic test facility.

Organization:

W. Kropp, Chalmers University of Technology, Division of Applied Acoustics C. Hoever, Continental Reifen Deutschland GmbH

	The influence of tyre cavity resonances on the exterior noise Wolfgang Kropp [Abstract] Rolling noise due to tyre/road interaction is mostly explained by tyre vibrations. The contribution of air-pumping might be an important contributor for certain cases. This observation has often been based on evaluating the speed exponent observed in measurements of the radiated sound. However simulations show that in many cases the observed speed exponent can be explained by tyre vibrations. In addition simulation show that for rough roads the resonances in the tyre cavity can play an important role. In the paper the influence of sound generation in the tyre cavity is discussed in more detailed. For this the Chalmers Tyre/Road Interaction model is utilised. The simulations clearly show the contribution of tyre cavity resonances to rolling noise as function of surface type. The results indicate that additional damping in he cavity would not only be beneficial for the interior noise inside the vehicle but in many cases also for the exterior rolling noise due to tyre/road interaction.
	Prediction of dynamic hub forces as a source of structure-borne tire/coarse road noise using a high-fidelity simulation approach Daniel De Gregoriis, Frank Naets, Peter Kindt, Wim Desmet [Abstract] When optimizing the tire/road structure-borne interior noise, multiple other coupled tire performances need to be simultaneously optimized. Typically, different tire design prototypes are built and tested. However, with the recent advent in virtual product development and the Digital Twin concept, high-fidelity numerical simulation approaches become a viable alternative to the experimental approach. These numerical approaches allow for e.g. design space exploration without the need for building and testing physical tire prototypes. Therefore, in this work the use of a fully predictive nonlinear numerical approach for the prediction of the dynamic contact- and hub forces of a tire rolling over a coarse road surface is described. These predicted hub forces can then be further used to assess the tire/road structure-borne noise performance of a tire. The influence of incorporating tire model nonlinearities as well as the air cavity and rim flexibility on the simulated hub forces is discussed. Furthermore, nonlinear model order reduction and hyper-reduction techniques are applied to greatly reduce the numerical simulation time, being approximately ten times faster than the experimental approach. Finally, a comparison between the predicted hub forces and experimental hub forces is shown, where a sufficient correspondence between both can be observed for design engineering purposes.
	Use of waste and marginal materials for silent roads Lily Poulikakos, Sahand Athari, Peter Mikhailenko, Zhengyin Piao, Muhammad Rafiq Kakar, Moises Bueno, Reto Pieren, Kurt Heutschi [Abstract] Inhabitants of the urban environment are exposed to high levels of traffic noise and increase in amounts of waste. One way to mitigate their exposure to noise is to take advantage of the existing public spaces between buildings and transform the available surfaces of pavements into elements that foster comfort and health for urban residents. This work aims to develop low noise pavements by using certain waste materials. Semi-dense asphalt (SDA) mixtures (Air Voids = 12-16%) will be used in this study due to their acoustic properties. Parts of the aggregate matrix will incorporate i) crumb rubber (CR) and ii) recycled concrete aggregates (RCA). The various mixtures produced will be compacted into 100mm dimeter specimens using the gyratory compactor which will be subject to texture measurements by 3D laser profilometry, porosity measure- ments and sound absorption measurements by the impedance tube method (EN 10534-2). This study will give us an understanding on how waste materials impact the texture and acoustic properties of roads and ultimately, to what degree they can be incorporated in low-noise asphalt mixtures.
	A LIFE NEREiDE test track with a poro-elastic pavement in Belgium Luc Goubert [Abstract] A poro-elastic road surface (PERS) is an innovative pavement consisting of stone aggregates and rubber granulates, bound with an elastic resin. It was invented around 1978 in Sweden and can yield unequalled reductions of the tyre/road noise: between 7 and 12 dB, whereas the best performing ”conventional” noise reducing pavement, two-layer porous asphalt, reaches maximum 7 dB. Since the invention of PERS, in several countries experiments have been carried out demonstrating the major drawback with this pavement type: its vulnerability and limited lifetime. In the FP7 PERSUADE project (2009-2015) a holistic approach was followed covering all relevant aspects. Several field tests on trafficked roads revealed that the prefab PERS slabs, glued on a rigid under layer is the most promising PERS technology. The LIFE project NEREiDE (2016-2020) focuses on the use of recycled materials in noise reducing pavements and comprises a work package about PERS. In the frame of the LIFE NEREiDE project, a 44 m long test track was constructed on a road in Gent, Belgium with prefab poro-elastic slabs, building on the technology developed in the PERSUADE project. This contribution outlines the construction process and the first test results obtained in the subsequent monitoring program.
	CFD Modeling of Pressure Variation in a Road Cavity with Volume Variation Marianne Bou Leba Bassil, Julien Cesbron, Philippe Klein [Abstract] This paper deals with the variation of air volume at the contact interface between a rolling tyre and a road surface leading to aerodynamic phenomena such as air-pumping. A numerical model was developed (in 2D and 3D) based on Computational Fluid Dynamics, using the Fluent solver, coupled with a displacement function of the bottom of the cavity representing its volume variation. Results showed that the maximum pressure reached at the bottom of the cavity during its complete closure increased as its volume decreased. The pressure variation was related to the volume variation by a relationship close to Laplace law for perfect gases. In addition, pressure oscillations after the opening of the cavity increased in amplitude and frequency as its volume decreased. Moreover, in the case of INRETS (1990) experimental setup, introducing the volume variation estimated by Laplace Law in the 3D model has shown better agreement between the calculated pressure and the measured pressure in comparison with the model neglecting this volume variation. Future work would therefore be to obtain the actual volume variation caused by the penetration of the tyre tread inside the cavity. This could be done either experimentally or numerically using a tyre/road contact model.
	Parametric Study of the Estimation of Indoor Trolley Wheel Stiffness for Use in a Rolling Noise Prediction Model Matt Edwards, Fabien Chevillotte, François-Xavier Bécot, Luc Jaouen, Nicolas Totaro [Abstract] In vehicle tyre/road contact modeling, dynamic models are typically used which incorporate the vehicle’s suspension in their estimation: thus relying on a known stiffness to determine the movement of the wheel in response to roughness excitation. For the case of a wheeled device rolling on a floor (such as would be the case for a delivery trolley moving merchandise around inside a commercial building), there is often no suspension, yet the wheel is still too soft to able to be considered mechanically rigid (as is the case in train/rail contact). A model which is aimed at incorporating the dynamic effects of the trolley in predicting the sound generated by rolling needs to provide a robust way of estimating the wheel’s effective stiffness. This work presents an original technique for estimating the stiffness of an indoor trolley wheel. A parametric study was conducted in order to identify the dependence of the wheel stiffness on each of the relevant variables: including the wheel’s radius, axle size, width, applied load, and material properties. The methodology may be used to estimate the stiffness of new wheel types (i.e. different geometries and materials) without needing to solve a finite element model each time.

Organization:

W. Kropp, Chalmers University of Technology, Division of Applied Acoustics J.S. Bolton, Herrick Laboratories/Purdue University T. Hashimoto, Seikii University

	Asphalt-surface defects detection, based on Tyre/Road noise analysis and geo-processing Carlos Andrés Ramos Romero, César Asensio [Abstract] A new approach to detect different asphalt defectology based on tyre/road signal-noise analysis by Machine Learning algorithms is proposed. The method exploits the acoustic signal generated by the rolling at the tyre-asphalt proximity zone. The audio data is recorded by a sensored vehicle that travelled an established route with supervised surface conditions. The acoustic data have shown to be relevant on automatic asphalt-defect classification by features extraction in frequency domain. In this study, local and distributed road-surface defects were classified using supervised pattern recognition algorithms. This new approach does not require specific on-board instrumentation, so any vehicle can easily become an asphalt status tester with the installation of a microphone.
	Investigation on Abnormal Gear Rattle Noise in Automated Manual Transmission (AMT) after Gear Upshifting Abhishek Lakhanlal Vaishya, Archan Sunilbhai Pujara, Devendra Kumar Khare, Kingshuk Satpathy, Vivek Singh [Abstract] Customers are getting delighted after receiving best in class vehicle NVH performance. In order to enhance customer NVH comfort, a specific abnormal gear rattle noise phenomenon during product development was investigated. The transmission produces gear rattling noise by the impacting of gears oscillating through their gear backlash. Similar noise was instigated immediately after gear upshifting, sustains for some time and then inaudible. In this case, the hearing frequency band is found higher than the usual gear rattling frequency band. Along with the assessment methodology, root causes of this abnormal gear rattle noise were discussed in this paper.
	Study for Whoop Noise Reduction in passenger vehicle with Clutch Hydraulic System Devendra Kumar Khare, Anant Saran Pandey, Abhishek Lakhanlal Vaishya [Abstract] The noise, vibration and harshness (NVH) is an important aspect in driving comfort improvement. Low noise and vibration is one of the primary requirements of passenger cars which play an important role in the customer’s perception. It includes a silent cabin with minimum vibrations and quite operation of the parts. In a vehicle, engine vibrations are transferred to clutch pedal through fluid pulsation inside the clutch operating system. In some cases, vibration gets amplified while operating clutch and a noise is generated which is called a whoop noise. Sound pressure level of whoop noise should be low to enhance customer comfort. The biggest challenge for NVH engineers in present scenario is to strike a balance between cost and good NVH. This paper elaborates the study of root cause analysis and reduction in whoop noise generated during clutch pedal operation.
	A Study of the Interaction Between Vehicle Exterior Noise Emissions and Vehicle Energy Demands for Different Drive Cycles Johan Nygren, Susann Boij, Romain Rumpler, Ciarán O'Reilly [Abstract] This initial study investigates the interaction between vehicle noise emissions and the energy required to move the vehicles along different drive cycles. There is often a trade-off between reducing noise emissions and at the same time reducing other environmental impacts. A vehicle’s energy demand associated with a specific drive cycle may be affected when a different route is chosen between two locations to reduce the noise exposure at certain observer points. The methodology of the study was to use the existing IMAGINE traffic noise model as a source model, and to evaluate the sound exposure level (SEL) at observer points as a function of instantaneous sound pressure level estimates as the source moved from one location to another via two different routes. A noise impact estimate with a linear dependence on the difference between the SEL and a threshold level was proposed. Also, the energy demand for each route was calculated. The results indicated that there is a difference between the two routes if the aim is to reduce the noise exposure or the energy demand. Suggested future research is to further improve the noise impact evaluations in the context of very short durations of exposure.
	Development of a Thermally Insulating Vibration Damping Compound Alexander Rasa [Abstract] Attenuating vibration in structures, such as marine vessels and rail carriages, through the use of damping materials can return a wide range of benefits from reducing air-borne and structure-borne noise to improving component life. Thermal insulation is typically also installed alongside vibration damping materials, preventing condensation and the transfer of thermal energy, therefore improving energy efficiency and the comfort of passengers and crew, along with increasing component life through the reduction of moisture. This paper details the development and evaluation of a single material, Decicoat T35, which is both vibration damping and thermally insulating, allowing easier and faster treatment over traditionally installing multiple materials. The compound is water-based, non-toxic and easily sprayable. The material has been evaluated to EN 12664 for thermal conductivity and to an internally developed method for anti-condensation in rail and marine industries. Decicoat T35 was also investigated using a variety of methods for vibration damping, including dynamic mechanical analysis (DMA) to ISO 6721-5 and experimental modal analysis (EMA). Furthermore, vibration damping performance was compared to a similar competitors sprayable thermal compound, highlighting the elevated vibration damping properties of the newly developed material. A thermally insulating vibration damping compound was successfully developed and evaluated.

Organization:

T. Van Renterghem, Ghent University, Department of Information Technology, WAVES research group

	Modelling reflections from single trees and entire forests Jean Marc Wunderli [Abstract] A model for the prediction of diffuse reflections from forested areas is presented. A single tree is thereby modelled as a cylinder of finite height, representing the trunk, and a number of spheres to account for reflections from the crown. The scattering properties of the involved structures are derived from the classical analytical solutions, complemented by a Fresnel zone weighting. For performance reasons not every tree in a forest is modelled individually but representative super trees are distributed over forested areas with a finer local resolution along the forest edges and a coarser representation in the depth of the forest. The model is formulated in one-third octave bands and yields sound exposure level as well as maximum sound pressure level. As the depth of the forest is automatically scaled correctly it can be applied for small-scaled vegetation and is for example capable of predicting levels behind forested barriers or berms. Comparisons with measurements showed a generally good agreement.
	Applying the diffusion equation to urban scenarios: Computational analysis of the diffusion coefficient Raúl Pagán Muñoz, Juan Miguel Navarro Ruiz, Maarten Hornikx [Abstract] Prediction of sound propagation is playing a key role in the planning and design process of urban areas and during the last decades different techniques have been developed for the computation of the sound fields in cities. Among them, the diffusion equation, based on the propagation of sound energy, is a simple and attractive tool for certain scenarios where the diffuse field is predominant, such as inner city environments. The diffusion equation is a well-known and efficient method to compute sound fields in rooms with a low amount of absorption. However, this method, which includes the diffusion coefficient describing the diffusivity of the energy propagation, requires further understanding for urban environments. In this work, the diffusion coefficient has been evaluated using wave-based numerical techniques for simplified two-dimensional urban scenarios. The parameter is computed spatially from the acoustic intensity and sound energy density fields in every direction for different geometries. This paper investigates the directional components of the diffusion coefficient.
	Solution of wide-angle parabolic equations for long-range sound propagation in a moving medium David Keith Wilson, Michael Muhlestein, Vladimir Ostashev, Michael J. Shaw, Michelle Swearingen, Sarah McComas [Abstract] Narrow-angle parabolic equations, which are widely used for outdoor sound propagation, are suitable for propagation angles up to 15∘-20∘ off the main axis (generally the horizontal direction). Wide-angle parabolic equations (WAPEs) are thus needed to accurately solve long-range problems involving refraction and scattering from elevated layers, such as audible- range propagation in stable nighttime conditions, and infrasound refraction by the stratosphere. However, it is difficult to derive WAPEs that are numerically feasible to solve while properly accounting for motion in the propagation medium (i.e., wind in the atmosphere). As a starting point, we adopt here a recently derived extra-wide-angle parabolic equation (EWAPE) for moving media which accounts for the propagation angles up to 90∘ with respect to the nominal propagation direction. A WAPE is then derived from the EWAPE using a Padé (2,2) approximation, which is valid for angles up to roughly 55∘. The resulting equation is generally suitable for long-range sound propagation applications in a moving atmosphere. We consider here solutions for example problems involving ducting and shadow zones. The solutions are compared to narrow-angle approximations and to wide-angle solutions based on the effective sound-speed approximation.
	Uncertainty Analysis of Environmental Sound : Analysis of a Series of Field Experiments John Fenlon [Abstract] A series of experiments used attended and unattended instruments at two primary sites: (i) adjacent to the A428 (a second-tier road), and (ii) adjacent to the West Coast mainline railway. These are both rural sites, so that the traffic / rail noise will tend to dominate even at some distance from the noise source. Various configurations of monitors at different distances from the noise source, on different days and for different times were tested, and sound measurements were integrated over 5-minute (road) or 1-minute (rail) periods. At each site a series of measurements were taken during July and August in two separate years, which included several 4-day sets, using two different types of noise meter. The underlying aim of the project was to develop a simple ëuncertainty’ budget for the measurement of environmental noise. The primary interest was in determining day-to-day variation, differences between instruments (at the same position), and differences between measurements at different distances from the source. One set of experiments also tested the difference between measurement positions chosen by different engineers.
	Model Test Stand for Acoustic Scattering Wolfram Bartolomaeus [Abstract] Scattering, diffraction and reflection on acoustical hard or soft obstacles affect the sound field in its propagation. The scattering of sound waves on an acoustical hard cylinder was calculated analytically using a harmonic development according to Bessel functions, simulated with the finite element method (FEM) and measured in the Laboratory for Acoustic Modeling on a scaled level. A good agreement between all three results was found. Similarly, the diffraction of a sound pulse at an edge was analytically calculated using the theory of Biot, Tolstoy and Medwin (BTM) and simulated with an FEM model. These simulations and the room impulse response measured in the Laboratory also agree satisfactorily. For further measurements, the new Acoustic Scattering Model Test Stand (MPaS) can be used also on objects or sets of objects that no longer can be described analytically in an easy way. Likewise, measurements at diffraction edges and reflection planes are feasible. These can be compared with analytical solutions or at least with FEM simulations. Thus, the MPaS test stand provides an important tool for modeling the propagation of the sound field over obstacles, which makes it possible in the future to construct e. g. optimized noise barrier attachments for roads.
	On Generation And Propagation Of Acoustic Surface Waves Over Rough, Periodic Surfaces Shahram Taherzadeh, Alex Stronach, David Berry, Keith Attenborough [Abstract] An acoustic surface wave is an evanescent wave that spreads cylindrically along the surface but decays exponentially with height above the surface. This wave travels at a little less than the speed of sound in air and is primarily related to the pore or surface. Surface waves are generated also when sound propagates over rough hard surface. This phenomenon is related to coherent multiple scattering from roughness elements. This paper investigates conditions under which surface waves are generated and propagated and the relationship between this surface wave frequency and the geometry are explored by measurements of the sound field due to a point source above periodically-spaced rigid strips placed on a rigid surface and by a Boundary Element Method in time-domain.
	A data bank of outdoor transfer functions Sylvain Cheinet, Loic Ehrhardt, Matthias Cosnefroy, Adrien Dagallier [Abstract] From the viewpoints of sound sensing and noise pollution applications, the propagation channel can be modeled by a transfer function. This transfer function is a sort of Grail, which readily connects any near-field signature to its propagated counterpart. In the scenario of outdoor propagation above a flat ground, the atmospheric conditions vary, and a data bank of transfer functions is needed. This study addresses the computational and physical challenges of forming such a data bank. It uses a three-dimensional, Finite-Difference, Time-Domain (FDTD) solver for propagating a reference wideband signal at 200 m above a finite-impedance ground. The simulations span over a wide range of meteorological conditions with consistent turbulence realizations. The formed transfer functions are compared to recent atmospheric-acoustic measurements. Their sensitivity to the wind direction and strength is discussed. The data bank is unique in its extensiveness and physical realism, and can be used for applicative purposes.
	Comparison of ASJ RTN-Model 2013 and the Harmonoise Engineering Model under Thick Barrier Configurations Takuya Oshima, Azusa Hoshikawa, Yumi Kurosaka [Abstract] Engineering models for environmental noise are developed in many countries under different backgrounds. Such differences make each model have different characteristics. In European countries, comparative studies have been carried out to illustrate the characteristics of models such as Nord2000, Harmonoise and CNOSSOS-EU. However, comparison of the Japanese road traffic noise model, ASJ RTN-Model 2013, to another model has not been made. The ASJ model uses A- weighted levels throughout the calculation process, unlike European models that use either octave-band or 1/3-octave- band levels. In this study, a methodology to derive consistent quantities between the ASJ model and the European Harmonoise model, such as the single-vehicle A-weighted power level, is established to make strict comparisons between the two models. The methodology is applied to thick barrier test cases that are included in a Harmonoise deliverable. At the same time, finite-difference time-domain simulations are carried out to obtain reference solutions. A-weighted source strengths between the two models are found to be larger for a light vehicle of the ASJ model. Comparison between each model and the reference solutions weighted by each source spectrum agree well for both models. The final A-weighted levels obtained by both models agree within a 2 dB difference.
	Managing the Uncertainty of Long-distance Sound Propagation from a Large Industrial Noise Source Tim Procter, Deanna Tomerini, Alan Lex Brown [Abstract] The attenuation of a sound signal over long distances is accounted for by atmospheric absorption, atmospheric stability, ground effects, barriers and spherical spreading. Conversely, the propagation of sound over long distances can be enhanced by source to receiver gradient winds and vertical temperature profiles that create downward refraction of the sound waves. While the nature of each of these phenomena is well understood the prediction of the sound propagation over long distances from a large industrial source such as an open cut mine is inconsistent with the measured sound pressure level attributed to the source. The uncertainty around the theoretical model predictions and the measured source sound pressure levels in a multi-source environment creates a control problem when managing day to day mining operations. This paper investigates the relationship between the measured and predicted sound pressure levels at three continuous noise meters along a propagation path at distances of 500m, 1500m and 2500m from the source. The objective of the analysis was to reduce the uncertainty in the measurement of the source’s contribution at the noise monitor located in a multi-source environment 2500m from source.
	Selection of the correction due to receiver location when performing environmental noise measurements using microphone array Luka Čurović [Abstract] Measurement and assessment of environmental noise is in many cases performed using procedures described in ISO 1996 - 2:2017 standard. As the measurement uncertainty is as important as the measurement result the standard gives guidance on the calculation of measurement uncertainty which is compliance with ISO/IEC Guide 98-3 (GUM). The sound pressure level is normally considered as the free field sound pressure level. But the measurement used in regulations are often performed in front of reflecting surfaces. In such cases one of the input quantities that determines the measured value is an input quantity to allow for any uncertainty due to selection of receiver location (δ_loc) that can take values from 0 to 6 dB. δ_loc has a profound effect on the measurement result. The choice of the right value is in many cases not trivial and only defined for traffic noise. In our study we will present measurements performed with a microphone array that can help in selecting the optimal measurement location when performing the noise measurement due to industrial sources in front of reflecting surfaces.
	Propagation of Bird Vocalizations in the Alpine Environment Didier Dragna, Loïc Berger, Sébastien Ollivier, Frédéric Sèbe [Abstract] Rock ptarmigan is a bird living in the mountains tops, whose population tends to decrease in Europe over the last decade, notably due to climate change, habitat loss and human disturbance. Censusing of number of birds in the Alps is realized via point-count monitoring using birds vocalization. However, detection of bird vocalizations by the human hear is somewhat tricky in these non-homogeneous landforms, due to sound propagation effects. In order to improve the counting method, it is necessary to understand how bird vocalizations propagate in the Alpine environment. With this aim, a propagation model based on a ray-tracing approach is developed and includes topography, ground reflection and absorption and atmospheric refraction and absorption. The model is first validated in both frequency and time domains against reference results obtained with a numerical solution of the linearized Euler equations for simple cases. Propagation of the bird vocalization in the Alpine environment is then illustrated. It is shown that the active space of the bird dramatically depends on the environment. The modeling of active and detection space will provide suitable information for birds monitoring, enabling managers to define more precisely their counting plans.
	Prediction of atmospheric sound propagation subject to parameter variability of atmospheric turbulence Jasmin Hörmeyer, Clemens Hübler, Tobias Bohne, Raimund Rolfes [Abstract] As part of the project ”WEA-Akzeptanz” an atmospheric sound propagation model will be developed to predict the sound immission from a wind turbine. The CNPE- (Crank Nicolson Parabolic Equation) method, based on the Helmholz equation is chosen as the sound propagation model. Complex environmental conditions such as atmospheric turbulence can be implemented with a random number generator in this model. Atmospheric turbulence is characterized by temporal fluctuating wind speed and temperature, which has an impact on the sound propagation and leads to scattered sound waves due to the turbulent eddies. Accordingly, the input of the model contains parameter variability. As a result, the sound pressure level at the point of immission also has a high variability and is not entirely deterministic. In this work, the influence of random input variables due to atmospheric turbulence on the uncertainty of the resulting sound pressure level will be investigated. Using this, the probability of the predicted sound pressure level can be quantified to get a better idea of sound immission under complex atmospheric conditions.

Organization:

F. Bertagnolio, DTU Wind Energy

In Situ Measured Facade Sound Insulation of Wind Turbine Sound Pontus Thorsson [Abstract] In most countries there are regulations of wind turbine sound level outdoors at dwellings. Often there are also regulations of the sound levels inside the dwelling, however not often directly aiming at wind turbine sound. The sound level indoors from wind turbines has attracted more interest in the latest years, and then especially in the low frequency region (up to 200 Hz). Studies on the in situ sound level difference between outside and inside of dwellings are however scarce. This paper presents the in situ measured sound level difference for two Swedish houses in rural locations, both using a loudspeaker and using the wind turbine sound as exciting signal. This is possible due to a 2 month long measurement series with simultaneous sound recordings outside and inside. The sound pressure level difference from the two methods are shown to differ substantially.

Prediction of Broadband Noise Generated from Turbulent Boundary Layers of a Horizontal Axis Wind Turbine Soichi Sasaki, Moe Htet Zaw [Abstract] We expand the blade element momentum theory for the prediction of the broadband noise of a horizontal axis wind turbine. For the prediction of the acoustic radiation from the turbulent boundary layers, the prediction theory suggested by B. D. Mugridge is applied. In this prediction methodology, the noise level of the wind turbine can be predicted by the model size of the isolated blade and the main dimensions of the objective wind turbine. At same time, the relative velocity and the attack angle becomes the important parameters. From the results of the wind tunnel test, NACA0018 generates the humped noise in the attached flow condition, whereas the noise spectra in the separated flow condition makes the broadband noise. We point out that the humped noise source in the wind turbine was made from the mid-span to the blade tip on the impeller based on this methodology. In the full length paper, the prediction methodology of the broadband noise, the experimental results of the wind tunnel test of NACA0018, moreover, not only the predicted wind turbine performance but the its broadband noise spectra generated from the turbine with 20m diameter will be presented.

Organization:

R.M. Alsina-Pagès, GTM - La Salle (URL) J.M. Navarro Ruiz, Universidad Católica San Antonio Murcia.

	Performance Analysis of the Acoustic Event Detector in the DYNAMAP's Rome suburban area Rosa Ma Alsina-Pagès, Francesc Alías, Joan Claudi Socoró, Ferran Orga [Abstract] Environmental noise is increasing year after year and, besides annoyance, it causes harmful health effects on people according to last 2018 WHO report. The Environmental Noise Directive 2002/49/EC (END) is the main instrument of the European Union to identify and combat noise pollution, followed by the CNOSSOS-EU methodological framework. In order to apply the END legislation, the EU Member States have to publish noise maps and action plans every five years. The use of Wireless Acoustic Sensor Networks (WASNs) changes the paradigm that addresses the END regulatory requirements as they enable the dynamic ubiquitous measurement of environmental noise pollution. Following the END, the DYNAMAP project develops a WASN-based low-cost noise mapping system to monitor in real-time the impact of road infrastructures in two pilot areas: Milan and Rome. To avoid biasing the noise maps with noise levels unrelated to traffic noise, an Anomalous Noise Event Detector (ANED) is included to remove them from the corresponding LAeq. The paper reflects the adaptation of the ANED algorithm to the WASN of the suburban area of Rome, which requires a specific analysis of the particularities of the suburban audio database, as well as future challenges and research on the generalization of the WASN.
	The Collation and Use of Data from Continuous Remote Monitoring Systems for the Control of Sound Emissions from a Large Industrial Noise Source Tim Procter, Stephen Lyons [Abstract] This paper shows how the analysis of the sound emissions from the constantly changing environs of a large open-cut mine using traditional closed-loop control theory identified the source of a time lag in the implementation of appropriate control strategies. As a result, a new control strategy based on ësmart’ technology has been developed to reduce the error signal time lag and help improve the control of noise emanating from the mine. Key elements of the new control system discussed in this paper include: the collation, analysis and reporting of the continuous real-time noise and meteorological monitoring data; confirmation of the source’s contribution in a multi-source environment; and identification of high-risk operational activities associated with the noise source. The smart technology presents this information in a format that: enhances the mining supervisor’s perception of the current environment; improves the comprehension of the data; reduces the uncertainty associated with identifying the mine’s contribution to the acoustic environment; and enables potential future actions and outcomes to be identified. This paper then shows that enhancing a user’s perception and awareness of the situation enables pre-emptive rather than reactive decision-making that results in reduced noise impacts and improved productivity.
	Estimation of Noise Immission Directivity using Small Microphone Array Jurij Prezelj, Luka Čurović, Tadej Novaković, Jure Murovec [Abstract] Environmental noise monitoring systems extensively use digital signal processing to perform computation and storage of noise levels, spectral filtering, narrowband spectral analysis, evaluation of statistical indices, wave recordings, detection of noise events based on thresholds, and other similar tasks. Increase in processing power can be exploited for development of advanced apparatus, capable of multichannel processing on hand held devices, as for example beamforming. A research has been undertaken to upgrade and implement beamforming for developing three new measurement options for environmental noise monitoring systems; 1) automatic exclusion of uncorrelated noise events from measurements, 2) automatic identification of the dominant noise source location and 3) automatic classification of observed noise event. Implementation of these three features can reduce need for human resources, resulting in reduced costs and more accurate results. Presented technical solution provides a synergy between these three options, implemented in an apparatus, which provides reliable results.
	Travel times in complex environments Adrien Dagallier, Sylvain Cheinet, Daniel Juvé, Aurélien Ponte, Jonathan Gula [Abstract] Times of Arrival (TOAs) of propagated signals are of utmost interest in seismic, underwater as well as aerial acoustics. From TOAs, one may reconstruct the propagation media properties from known sources and sensors positions, or conversely, find sound events locations in a known environment. Modeling the TOAs in realistic environments (wind or current, sound speed gradients, obstacles...) requires a very general physical model, able to factor in the impact of refractive and diffractive processes. We present an interface-tracking model based on Sethian’s Fast-Marching method for computing TOAs. This method is applicable to complex 3D meshed environments. Examples will be given in an urban area, in the complex atmosphere, as well as in the ocean. The combination of the marching model with the time-matching principle is illustrated to yield a general localization framework applicable to these various environments.

Organization:

G. Dutilleux, Norwegian University of Science and Technology

	Quality assurance methods demonstrated with the calculation of sound propagation with ISO 9613-2 and with CNOSSOS-EU Wolfgang Probst [Abstract] Many methods have been implemented in legally binding rules how sound propagation shall be calculated to support the minimization of environmental noise. With the series of the International Standard ISO 17534 we created a framework to ensure this necessary precision. One of the main pillar of the strategy is to clarify all ambiguous definitions that will cause different results if programmed by different software developers. With the example ISO 9613-2 examples are shown where even existing definitions had to be modified. It can be stated that all calculation methods have their shortcomings that can only be detected if the method is realized in software and then practically applied. The same is true with the harmonized European calculation method - here abbreviated as ”CNOSSOS-EU”. Examples are shown where it was necessary to modify and clarify some definitions before it was possible to create test cases to check the correct implementation of the method. With this work on about 30 Test Cases for ISO 9613-2 and for CNOSSOS-EU we learned a lot about the optimal strategy if a new method shall be implemented or an existing method shall be improved. Some recommendations for future projects of that type will be given.
	Meta-Modeling for urban noise mapping Antoine Lesieur, Pierre Aumond, Vivien Mallet, Arnaud Can [Abstract] Urban noise maps are usually computed by noise propagation models coupled with emission and attenuation rules like CNOSSOS. These models can require several hours to produce a map over a city for a single set of input data. This computational cost makes the models unusable for applications like uncertainty quantification where hundreds of simulations may be required. One solution is to replace the physical model with a metamodel which is extremely fast and yet fairly reproduces the results of the physical model. The strategy is first to reduce the dimension of both inputs and outputs of the physical model, which leads to a reduced model. This reduced model is then replaced by a statistical emulator. The emulator is trained with calls to the reduced model for a set of chosen inputs. The emulator uses linear regression and interpolation between the training output values. We applied this approach to the NoiseModelling software over part of Paris, using either radial basis function or Kriging at the interpolation step. It required a training set of 1000 calls to the physical model. The metamodel is 10000 times faster, and preserves the main behavior of the physical model, with only 1dBA of quadratic error.
	Global sensitivity analysis for urban noise modelling Pierre Aumond, Arnaud Can, Vivien Mallet, Benoît Gauvreau, Gwenaël Guillaume [Abstract] Regulatory noise cartography relies on many input data that can be incomplete, erroneous or inexistent. Thus, operators have to complete/validate/qualify those data by prioritizing some information sources and parameters to the detriment of others. Consequently, the sensitivity of the noise model to such input parameters should be determined to focus on the most influential factors. For this purpose, a global sensitivity analysis was carried out on the open-source software NoiseModelling, which integrates the CNOSSOS regulatory method, using the Morris screening method. As part of the ANR CENSE project (2016-2020), a global sensitivity analysis will be presented on a case study of a 1 km2 area in the city of Lorient in France.
	Estimating of dose-responce relationship of Shinkansen railway noise using noise mapping Yasuhiro Hiraguri, Akinori Fukushima, Takashi Morihara, Shigenori Yokoshima [Abstract] Up until now, in Japan, the dose-responce relationship of Shinkansen railway noise has been estimated using measurement-based noise levels as dose value. In this paper, we newly propose a noise mapping technique of Shinkansen- railway and re-estimate the dose-responce relationship of Shinkansen-railway noise.
	Uncertainty analysis for the in situ sound power level determination using the substitution method Spyros Brezas, Volker Wittstock [Abstract] Sound power determination at different environments and using different techniques exhibits deviations especially at low frequencies. In previous studies the establishment of traceability has been introduced and the realisation and dissemination of the unit watt in airborne sound has been discussed. For the completion of traceability, the sound power level determination of realistic sources is required. The contribution presents the application of the substitution method for the in situ sound power determination of real sources. The substitution method includes both sound pressure and sound intensity measurements. For collecting experimental data, different measurement methods were applied in various environments. Following the concept of traceability, the sound power level is accompanied by the related uncertainty, which explicitly explains the contributing factors along with their extent of contribution. A comparison is made between the uncertainty levels using sound pressure and sound intensity. The proposed uncertainty covers a wider frequency range compared to the up-to-date standards.

Organization:

J. Forssén, Chalmers University of Technology, Architecture and Civil Engineering E. Zea, KTH Royal Institute of Technology

	Future Low-Noise Aircraft Technologies and Procedures – Perception-based Evaluation using Auralised Flyovers Reto Pieren, Lothar Bertsch, Demian Lauper, Beat Schäffer [Abstract] Residents living in the vicinity of airports are exposed to noise from departing and approaching aircraft. Noise may be reduced by introducing novel aircraft technologies and flight procedures. By means of auralisation and listening experiments, such possible future changes can be evaluated by considering human sound perception. In this study, flyovers of different aircraft types and flight procedures were auralised at multiple virtual observer locations, and subsequently evaluated in a psychoacoustic laboratory experiment with respect to short-term noise annoyance. Approaches of an existing reference aircraft, a possible low-noise retrofitted version and a future low-noise design were simulated with standard and tailored procedures. Separate source signals were synthesised for engine broadband, fan tone, airframe broadband, and cavity tone noise. Further, smooth transitions between configurations and operational conditions were modelled to create realistic sounds. To increase plausibility, the propagation simulation considered ground reflection and frequency-dependent amplitude modulation due to propagation through a turbulent atmosphere. The flyover sounds were spatially reproduced by a hemispherical loudspeaker array. The listening experiment revealed significant annoyance reductions for low-noise aircraft types and flight procedures, that maximal benefit is achieved by the combined optimization of aircraft technology and procedure, and that distributed observers need to be considered.
	Auralization of Aircraft Noise in an Urban Environment Roalt Aalmoes, Maurits Van Der Veen, Henk Lania [Abstract] Realistic simulation of aircraft fly-over using auralization and visualisation techniques help in the research towards more silent aircraft types and/or operations. Up until recently most auralizations of aircraft were created in a virtual countryside, only taking the direct path and ground reflection into account. In an urban environment, reflections and absorption of buildings must be considered as well as different ambient sound levels. This research effort aims at creating an auralization of a turbofan aircraft flying over an urban environment. Fly-over recordings were obtained of aircraft approaching Amsterdam Airport Schiphol. The spectral content of the aircraft noise is used to synthesize the source noise of the aircraft. A dynamic 3D model of the urban environment is constructed using publicly available city data. This data is combined with recorded video data to match visual and audio occlusion of the aircraft by buildings. Reflection surfaces are used to determine mirrored noise sources, which are spatially added to the 3D buildings. At the observer the sound from the different paths is received and presented as binaural (directional) sound together with a realistic visualisation in Virtual Reality using the Oculus Rift.
	Physically-Based Auralization of Railway Rolling Noise Julien Maillard, Abbes Kacem, Nadine Martin, Baldrik Faure [Abstract] Railway noise contribute significantly to noise pollution both outside and within cities. In recent years, prediction models have been developed to study exposure levels and evaluate abatement solutions. Going one step further, auralization may provide an effective mean for evaluating perceptually the influence of railway noise to the soundscape near existing or future infrastructures. This paper extends railway noise emission models to propose an auralization approach based on physical parameters. As a first step, the approach focuses on rolling noise radiated by the track and wheels, which represents the dominant noise source over a wide range of speed. The excitation of the wheel/rail system by surface roughness is modeled in the time domain. The rail contribution is synthesized using a set of discrete coherent monopoles, preserving the characteristics of the emitted sound field. The contribution of the wheel uses resonant filters based on its structural response. Finally, the contribution of track sleepers is included following the standard TWINS model. Preliminary validations of the approach compare sound levels with existing models as well as measured data. Results from listening tests evaluating the realism of auralized pass-by noise samples are also presented.
	Recording-based Auralization of Train Noise in a Digital Urban Sound Planning Toolbox Krister Larsson, Jens Forssén, Peter Lundén, Xuetao Zhang, Andrea Sandberg [Abstract] Densification around railway stations has the potential to create more climate-neutral transport and resource-efficient travel, but the noise challenge is a limiting factor for sustainable and effective solutions. Currently, railway traffic increases in Europe and major investments in railway infrastructure are planned in Sweden and other European countries. The idea of this project is to finalise a demonstrator of a digital planning tool that facilitates investment in innovative noise measures on or in connection with railway infrastructure close to stations. The digital toolbox will provide the opportunity, at early planning stage, to virtualize the environment for common train types and traffic situations, to describe the effects of planned noise measures including auralization as well as socioeconomic costs and benefits, impact on property values and climate impact. The auralizations developed in this project are based on binaural and ambisonics recordings, which are modified to simulate several measures including rail dampers and various noise barriers such as thin and tall screens and low-height barriers of different widths with acoustically hard or soft surface materials as well as finite lengths. The paper describes the method of auralization, as well as the co-creation and workshop approaches used for the development of the toolbox.
	A Parametric Method to Synthesize Wind Turbine Sounds Pontus Thorsson [Abstract] The annoyance to wind turbine sound is today well known. There is no doubt that increased sound pressure level causes in general an increased annoyance. However, when it comes to additional descriptors of the wind turbine sound, such as e g equivalent frequency spectrum or amplitude modulation, there is no consensus on what parameters that are most important. In the research project Wind Turbine Noise Effects on Sleep (WiTNES) there was a need for 8 hour long sound signals from wind turbines that were completely free of extraneous sounds. Recordings from reality are difficult to maintain free from disturbances. Instead a parametric synthesis method was developed that is capable of creating wind turbine sounds of arbitrary length. The sounds can be designed in terms of equivalent frequency spectrum, what frequency bands that have amplitude modulation as well as frequency dependent amplitude modulation strength and variation. Recordings of wind turbine sounds at immission distances from different wind turbine types were used to find relevant values for the parameters. This paper describes the synthesis method and some examples of input parameters.
	Analysing the Effectiveness of Approaches to Auralisation for Applications in Environmental Acoustics Alex Southern, Frank Stevens, Damian Murphy [Abstract] Auralisation is becoming more commonplace as a means of communicating and disseminating plans for changes in our environmental landscape due to major infrastructure projects such as road, rail and airport developments, where noise can be a significant issue. This paper presents an approach to this challenge in the context of a major public consultation for a UK road project, and offers comment on different means for facilitating presentation of, and engagement with, both audio and visual material. Although such methods are generally used for information only, every effort is made to ensure that the final result is both plausible and comparable to existing on-site conditions, and calibrated accordingly, with a view to supporting other more standardised methods and metrics. However, it is also possible to develop more quantitative measures as to how effective and successful such auralisations are in terms of conveying a representation of a given scene, and any planned for interventions. In this work the Self Assessment Manikin and soundscape classification descriptors are used in a series of listening tests to better understand the approach that has been adopted for this project.
	Auralisations for Outdoor Noise Sources Per Finne, Erik Thysell, Christian Weirum Claumarch [Abstract] For years auralisations have been used in addition to mandatory noise maps and calculated number of noise exposed dwellings in the EAA processes carried out by i.g. the Danish Road Administration. Auralisations make it possible to get a realistic impression of the future noise during the planning of new roads or other important noise sources. Till now, only major road projects have been published using auralisations. Noise sources like wind turbines, railways and light rails, can obviously benefit from auralisations in communicating noise consequences to the influenced people influenced like neighbours etc. Auralisations makes it possible to aurally explain the mix of noise combined by two or even more noise sources. Normally, noise from various noise sources are treated and explained individually, and they even have different noise limits to meet. Auralisations are made to represent the total soundscape which is much more realistic to the noise in the close surroundings as it will be perceived and experienced by the future neighbours. This presentation deals with auralisations made for other sources than road noise and the combined noise generated by various noise sources. The project is supported by the Danish Agency for Institutions and Educational Grants.
	Audiovisual simulation inside the residential rooms of roadside buildings Asakura Takumi, Riku Hashimoto [Abstract] This study aims at an audiovisual simulation scheme of the environment inside such a room inside roadside residential buildings, which can be applied to the physical/physiological evaluation of the environments inside unbuilt buildings. For that purpose, in this study, the authors performed a basic study for composing a simulation scheme of both the sound and visual environment inside rooms. In the study, the sound of the passing vehicles transmitted into the rooms were synthesized based on measured dry sources of the pass-by vehicle sounds, whereas the 3-dimensional visual settings inside the room were made by using the 3-dimensional modeling software of Unity. In addition, by using the method, the subjective evaluation experiment on the impression of the audiovisual environment inside the virtual rooms were performed by using the synthesized sound and visual data. Based on the results of the experiment, the subjective effects caused by the auditory and visual stimuli were discussed.
	Traffic Flow Auralisation based on Single Vehicle Pass-by Noise Synthesis Yang Fu, Damian Murphy, Alex Southern [Abstract] Research into the auralisation of road traffic noise has drawn more attention in the recent years, as it can provide a more direct and immersive experience with which to communicate the implication of major urban design decisions with city planners and the public. Some synthesis models of single vehicle pass-by noise, and methods for outdoor sound propagation modelling have been proposed, which may be integrated for auralisation of traffic flow scenes. In this paper, a framework for traffic flow auralisation is presented. The synthesis model is composed of multiple single vehicle pass-by examples, based on tyre noise and engine noise synthesis. A first-order image source method is implemented, as a first approximation to represent sound propagation across a local region of a wider urban environment. The resulting plausibility of the proposed traffic flow auralisation model under different traffic flow scenes is tested using both objective sound quality metrics and subjective listening tests.
	Progressive region-of-interest filtering for urban sound auralization applications with multiple reflected and diffracted propagation paths Jonas Stienen, Michael Vorländer [Abstract] When auralizing urban environments, two major challenges can be identified for the propagation simulation: the extent of the built environment and the wave-based propagation features, above all diffraction into the shadow region. It is in the nature of open space that sound can propagate long distances. For the auralization of city acoustics without a-priori knowledge on the sound source characteristics huge data must be taken into account. With the increase of freely accessible open format databases providing layers with simplified geometric models of entire landscapes including buildings, a solution to create virtual acoustic environments based on this data is desirable. In this attempt, a progressive broad-band propagation simulation model based on Geometrical Acoustics is proposed that primarily exploits the frequency-independent energy loss subject to geometrical spreading of spherical waves for the spatial filtering of the geometry input. An ellipsoidal hull representing a region of interest is constructed from the source and receiver position as well as sound power and sensitivity, respectively. Subsequently a fast filtering algorithm with linear complexity removes approximately irrelevant geometry for each propagation step releasing the propagation algorithm from executing ray-polygon intersection tests on the entire geometry dataset and speeding up the propagation simulation significantly.
	Sound insulation auralization filters design for outdoor moving sources Imran Muhammad, Anne Heimes, Michael Vorländer [Abstract] Auralization techniques are rapidly attaining popularity, mostly due to their applications in building acoustics and virtual reality. As an application, sound insulation auralization has become a valuable tool to assess the sound insulation of buildings in terms of investigating how the indoor and outdoor noise sources are influencing the daily working and living routine of humans. The airborne sound insulation prediction standards have several simplifications that are implicit in the formulation on which they are based. Therefore, it is necessary to develop a model to build up an auralization chain of the sound insulation by taking into account the complex wave fields on the surfaces of the building elements and the spatial and temporal variation of the sound pressure field inside arbitrarily shaped rooms. We investigate the bending (flexural) wave patterns on the extended walls of buildings, the directionality of the outdoor moving sound sources to construct perceptually correct binaural filters. Additionally, we address the problem of synthesising the room impulse response from the one-third octave band values of the reverberation times. Our proposed approach increases acoustic interaction with the virtual environments creating a deeper, more realistic and immersive scene and leading to more accurate perceptual evaluations of sound insulation.

Organization:

K. Attenborough, The Open University

	A review of natural means for noise abatement Keith Attenborough [Abstract] In addition to wavefront-spreading and air absorption, noise from sources near the ground can be abated during its path to the receiver. ëNatural’ ways of doing this are described. While the reduction in sound levels attributable to ground effect, sometimes called ground absorption, is included in prediction schemes for surface transport noise, it is possible to increase it either by choosing particular types of acoustically-soft ground surface or by deliberately introducing roughness on an acoustically-hard ground. The most effective acoustically-soft grounds are those with low flow resistivity and recent cultivation without compaction. Low parallel walls or lattice arrangements have been shown to be useful forms of roughness. Reflections from façades and barriers can be reduced by vegetative coverings. While hedges offer relatively little noise reduction, 15 m wide belts of trees can be designed to give on the order of 10 dB reduction at a receiver 30 m from a road. Potential noise reductions from combining ënatural’ methods in situations where barriers are not feasible will be explored.
	The influence of organic matter on acoustical properties of soil Kamrun Suravi, Ho-Chul Shin, Keith Attenborough, Shahram Taherzadeh, Richard Whalley [Abstract] To study the effects of organic matter on the physical and mechanical properties of soils, which may influence the propagation of sound across cultivated land and acoustic-to- seismic coupling, soil samples from a long-term agriculture experiment involving different treatments, have been subjected to a triaxial test method to measure the normal consolidation and shear deformation. Also, differential thermal analysis (DTA) has been used to deduce the organic matter content and its composition and water release characteristics and soil strength have been measured with the suction plate method and indirect tension method, respectively. Data show the main effect of organic carbon is on the porosity at any consolidation pressure and that porosity in soil, shear deformed to the critical state condition, is correlated with internal surface area inferred from DTA. Indirect tension shows significant variation in the tensile strength (F= 8.95; P<0.001) of soils from different management practices. Initial acoustic impedance tube measurements made on sand samples with various water content show similar results to those made by Horoshenkov and Mohamed (2006). Results of further measurements on soils with different organic matter content will be presented.
	Acoustic absorption of a living green wall - Parametric transducer and XYZ gantry measurement method Anna Romanova, Kirill Horoshenkov [Abstract] This work provides evidence of the importance of considering urban green structures and in particular the Living Green Walls (LGW) as sound-scaping instruments. An insight to Living Green Walls sound absorption properties is provided with particular interest of sound absorption coefficient and plant morphology. The work was carried out in-situ in internal and external environment and makes use of parametric transducer and XYZ gantry. Novel method of parametric transducer was previously verified by the use of impedance tube tests bench-marking against porous media material. The proposed XYZ gantry method eliminates the reflection from the ground and averages the discrepancy of the absorbed/reflected sound due to plant morphology over a given area of interest. The new method is an alternative to the ISO354-2003 and CEN/TS 1793-5:2016 standard methods to measure acoustic absorption of materials.
	Sound Propagation in a Forest Based on 3D Multiple Scattering Theories Vladimir Ostashev, David Keith Wilson, Michael Muhlestein, Keith Attenborough [Abstract] This presentation overviews recently developed approaches for forest acoustics based on three-dimensional (3D) multiple scattering theories. The mean sound field is calculated with 3D effective wavenumber which represents the scattering by trunks, branches, and the canopy. The radiative transfer equation is formulated appropriately to forest acoustics; this result enables calculations of the mean sound intensities transmitted and backscattered from a stand of trees. Using a modified Born approximation, pulse propagation in a forest is analyzed and compared with results of measurements. In addition it is demonstrated that the equations for the first two statistical moments of the sound field in a forest have the same form as those in a turbulent atmosphere if the scattering properties of the media are expressed in terms of the cross sections. This correspondence, and the existing theory of the interference of the direct and ground reflected waves in a turbulent atmosphere, enable analysis of a similar effect in a forest, where these waves are scattered by forest elements. The results obtained are compared with data.
	Sound Absorption by Tree Bark Mengmeng Li, Timothy Van Renterghem, Jian Kang, Dick Botteldooren [Abstract] Scattering of sound waves by trunks is part of the noise reducing potential of tree belts, and it has been shown before that the absorbing properties of the trunks are relevant in this respect. Information on bark absorption is currently very limited. Therefore, laboratory experiments were conducted with an impedance tube to measure the bark sound absorption of various species, including variation in bark thickness, tree age, bark porosity, density, and trunk diameter. Preliminary measurements were made to define the relevant part of the trunk for its acoustic absorption and to come to a sample handling procedure. From each trunk, 4 cylindrical samples were were made including 2 centimeters of wood behind the bark. The measurements show that the absorption (at normal incidence) is generally below 10 %, and that there is a small variety in between species.

Organization:

B. Xie, South China University of Technology P. Majdak, Acoustics Research Institute, Austrian Academy of Sciences

	Investigating Uncertainties in Fast HRTF Measurements Shaima'A Doma, Hark Braren, Janina Fels [Abstract] In personalized binaural synthesis, the use of individual head-related transfer functions (HRTFs) promises improved auditory immersion and localization performance, compared to generic HRTFs. A direct way of acquiring individual HRTFs is via acoustical measurements. At the Institute of Technical Acoustics of RWTH Aachen University, a measurement arc with 64 loudspeakers has been developed and optimized for this purpose, performing a fast measurement whilst rotating continuously around the subject. Optical tracking allows an investigation into positional uncertainties of the arc, as well as into subject movement. Previous evaluation approaches of the measurement system are expanded in this study. A series of human HRTF measurements is conducted, together with 3D structured light head scans of the subjects, which can be used for Boundary Element Method (BEM) simulation of the HRTF. By means of objective distance measures, differences between HRTFs, as well as repetition errors are quantified and measurement accuracy is assessed.
	Independent modes and dimensionality reduction of head- related transfer functions based on tensor decomposition Tong Zhao, Bosun Xie [Abstract] Head-related transfer functions (HRTFs) are acoustical transfer functions from a source to two ears in the free-field and vital to spatial audio. Generally, far-field HRTFs vary with source direction, frequency and subject. Accordingly, the dimensionality of a full set of HRTFs is very large. Linear decomposition has been used to reduce the dimensionality of HRTFs. Due to the multiple variable-dependent characteristics of HRTFs, in present work, tensor decomposition is applied to analyze the independent modes of HRTF variation so as to achieve a more efficient representation of them. The measured HRTF magnitudes are first smoothed with a frequency bandwidth of 1 ERB. Then the smoothed HRTF magnitudes of 52 Chinese subject, 493 directions for each subjects and 42 discrete frequencies (up to 14 kHz) are combined as a 3rd-order tensor and then undergone a Tucker decomposition. The results indicate that 8 subject-related modes, 13 directional-related modes and 6 frequency-related modes account for more than 95 % variations of HRTFs at corresponding dimension respectively, and thus the HRTF dataset can be reduced by truncating the modes to corresponding orders. Based on the results, it is also deduced that 8 anthropometric parameters at least are needed for the anthropometry- based HRTF customization.
	Objective Differences between Individual HRTF Datasets of Children and Adults Hark Braren, Janina Fels [Abstract] Subjective evaluations of noise by children and its effect on their learning abilities in classrooms are a current research topic. Studies in this field often make use of virtual acoustic scenarios to be able to present a controlled acoustic environment in which to conduct the experiments. The presentations of realistic acoustic scenes requires a set of Head-Related Transfer Functions (HRTFs) that in the best case is the subject’s individual HRTF, or matches it as closely as possible. However in most cases only a generic HRTF from an artificial head is available. This leads to perceptive deviations in location, coloration and reduces the immersion in the presented scene. Previous studies have shown, that anthropometric parameters, and by inference HRTFs differ especially for younger children compared to adults. In this study, individual high-resolution HRTF datasets as well as head and pinna geometries of 1st and 2nd grade children (ages 6-9 years) have been measured using the fast measurement system at the Dep. of Medical Acoustics (Institute of Technical Acoustics, RWTH Aachen University) and a structured light 3D Scanner. The HRTFs and anthropometric measurements are compared to those from adults and commonly available artificial heads to objectively evaluate their similarities and differences.
	Rendering virtual source at various distances using binaural Ambisonics scheme in dynamic virtual auditory display Bosun Xie, Jianliang Jiang [Abstract] In dynamic virtual auditory display, free-field virtual source at various directions and distances is conventionally realized by filtering the input stimulus with near-field HRTFs. However, this conventional method meets with some problems in practical use, due to the difficulty in acquiring near-field HRTFs and requiring a large memory to storage them. In present work, a scheme for rendering virtual source at various distances using binaural near-field compensated Ambisonics with appropriate order is proposed. Based on the principle of spherical Bessel and harmonics decomposition of sound field, the directional and distance information of sound field is first encoded into Ambisonics signals and then converted to headphone signals by dynamic binaural synthesis using far-field HRTFs only. It is proved that the 5th order dynamic binaural Ambisonics with 36-64 pairs of HRTF- based filters is able to recreate desired binaural pressures for various source directions and distances up to 2 - 3 kHz, covering the frequency range for ITD (and its dynamic variation) as well as low-frequency ILD as directional and distance localization cues, respectively. A psychoacoustic experiment validates that the proposed scheme is able to recreate appropriate distances localization perception outside the median plane.
	Local Ambisonics panning method for creating virtual source in the median plane Kailing Yi, Bosun Xie [Abstract] With the development of multichannel sound with height, it is desired to create virtual source in different vertical directions, especially in the frontal median plane. For this purpose, a local Ambisonics panning method is proposed. In a case of analysis, three loudspeakers are arranged in the frontal median plane, with one locating in the horizontal plane and other two locating 30∘-45∘ above or below the horizontal plane, respectively. The signal amplitudes of loudspeakers are a linear combination of the zero order and first order harmonics of elevation angle. It is proved that, by properly choosing the weights of linear combination for each loudspeaker signals, the current method is able to create appropriate interaural time difference as well as its dynamic variation caused by head rotation and head tilting. Therefore, based on classical Wallach’s hypothesis and its modern validations, the current method is able to create appropriate dynamic cue for vertical localization in the median plane at low frequency. Results of psychoacoustic experiment indicate that the current method yield reasonable localization performance and is superior to traditional pair-wise amplitude panning. The method can also be extended to create virtual source in other vertical planes.
	The limitation of static transaural reproduction with two frontal loudspeakers Lulu Liu, Bosun Xie [Abstract] Incorporated HRTF-based binaural synthesis and cross-talk cancellation, transaural reproduction with two frontal loudspeakers is theoretically able to duplicate the binaural pressures of an actual sound source, and then is expected to be able to recreate virtual source at arbitrary direction. However, many experiments have indicated that for static transaural reproduction, in which transaural processing is fixed regardless of head turning, the perceived virtual source is usually limited in the frontal-horizontal plane. The reason may be that, in static transaural reproduction, the dynamic cue for front-back and vertical localization is omitted, while the high-frequency spectral cue is unstable against head movement. To validate this hypothesis, a psychoacoustic experiment is conducted in present work. A dynamic transaural system with two frontal loudspeakers is used in the experiment. According to the contemporary orientation of head detected by a head-tracker, the system changes the transaural processing so that binaural pressures in reproduction follow the turning of subject’s head. The experimental results indicate that, by incorporating the dynamic cue caused by head turning, dynamic tranaural reproduction is able to recreate virtual source at front, back and vertical directions. Therefore, the limitation in conventional static transaural reproduction is indeed caused by the omitting of dynamic cue.
	Effect of individualized near-field head-related transfer functions on distance perception in dynamic virtual auditory display Guangzheng Yu, Liliang Wang, Liang Linda, Bosun Xie, Jianliang Jiang, Chengyun Zhang [Abstract] Virtual auditory displays (VAD) aim to create virtual source at various directions and distances in headphone reproduction by using HRTF-based binaural synthesis. Incorporating the variation of binaural signals caused by head turning in three degrees of freedom, a dynamic VAD reduce front-back and up down confusion, and improve the externalization of perceived virtual sources. It is well known that individualized far-field HRTFs also improve the direction localization of perceived source. In present work, a virtual source localization experiment is conducted to further examine the effect of individualized near-field HRTFs on direction and distance perception in dynamic virtual auditory display. The results indicate that dynamic binaural synthesis with both individualized and non-individualized near-field HRTFs is able to create perceived virtual source at various target distance within 1.0 m outside the median plane, especially at the lateral directions. Individualized HRTFs influence little to distance perception, but improve accuracy of directional perception in terms of elevation errors.
	3-D Sound Image Localization in Reproduction of 22.2 Multichannel Audio Based on Room Impulse Response Generation with Vector Composition Kaige Zheng, Misaki Otsuka, Takanobu Nishiura [Abstract] To simplify content creation processes and improve the realism of 22.2 multichannel system, a method to localize sound images is derived in this paper, which is an improvement of the method based on vector base amplitude panning. The sound image localization can be realized by a specific direction and distance. A sound field is simulated to make spatial impressions in 22.2 multichannel reproduction, which includes reflection and reverberation, and the simulation approach is based on room impulse response generation with vector composition. To control the sound pressure on the ears of the listener, the amplitude of the input signal is attenuated in advance by the distance of sound image. Evaluation experiments were carried out both subjectively and objectively with binaural recording. The improvement in the reproduction of sound image’s distance is realized, while the direction of the sound image stays the same as the method based on vector base amplitude panning.
	Evaluation of Moving Sound Image Localization for Reproduction of 22.2 Multichannel Audio Using Up-mix Algorithm Hiromu Suzuki, Takanobu Nishiura [Abstract] 22.2 multichannel audio had been developed for ultra-high definition televisions. It can reproduce a three-dimensional (3-D) sound field by simultaneously driving 22 loudspeakers and two woofers. However, it is difficult to easily acquire multichannel signals because of equipment costs and so on. The up-mix algorithm for generating multichannel signals from the existing fewer channel signals has been proposed to solve this problem, Whereas, the conventional up-mix algorithm has been proposed without the 3-D sound source location. As a result, it cannot reproduce the moving sound image in vertical and back-ward directions. In this paper, we propose a novel up-mix algorithm which extends two channel signals to 22.2 multichannel signals while using the 3-D location of the sound source. Both objective and subjective evaluations were performed to comparing with the accuracy of reproduced sound image localization. Results of the evaluations show that the proposed up-mix algorithm can reproduce 3-D moving sound images more accurately than the conventional one.
	[Abstract] Depending on the spatial audio reproduction method, surrounding spherical loudspeaker arrays enable either physically correct sound field synthesis in a physically constrained sweet spot or rather aim at perceptually correct reproduction. Feasible methods include modal-based approaches such as higher-order Ambisonics with a continuous virtual panning function, facilitating phantom sources with panning-invariant energy and spread by use of sub-band weighting for side lobe suppression and a suitable decoder strategy for the given loudspeaker layout. Instead, multiple-direction amplitude panning is based on three-dimensional amplitude panning with discrete loudspeaker gains and vectors, also aiming at constant spread of phantom sources with reduced coloration, but without claiming physical correctness. Representing a fundamentally different approach, acoustic crosstalk cancellation relies on the concept of binaural technology and enables the creation of physically correct signals at the entrance of the ear canals. Although these methods have been well researched individually, there has been only little comparative measurement-based evaluation to validate simulations considering practical imperfections. Such a comparison is presented based on various error metrics.
	Digital Cinema: Analysis of Multi-Channel Systems to Control Sound Levels Peter Houtave [Abstract] Since 2014, Flemish Authorities have installed an auto-regulatory system for cinemas in order to control the sound pressure levels during screening of different digital source materials. In preparation of this regulation, an acoustic study was necessary to fully understand the existing sound pressure levels and how they are generated in different movie theatres. Therefore, the whole sound system was studied in extensive details to describe the subsequent steps in the motion-picture multi-channel sound reproduction chains. The analysis started at the levels [dBFS] of the DCP 5.1 multichannel digital source signals, passing through the signal processing and amplification, integrating the electro-acoustic and room responses at different listening positions. As a result, it became clear that in modern digital cinemas with proper room calibration, a good prediction of the resulting sound pressure levels can be achieved. This result provides therefore a mean of control of the resulting sound pressure levels even before actual playback. Different sources [film, trailer, commercial announcement], different sizes of venues and different audience positions were investigated. Results of the different analysis steps will be presented and discussed.

	Estimation of Cross-Talk Compensation Filter using Bone Conduction Ear Microphone Irwansyah Irwansyah, Tsuyoshi Usagawa [Abstract] Placing a bone conduction (BC) transducer anywhere on the head will result in a BC sound reaching the cochlea in both ears, unlike air conduction. This “cross-talk” limits a listener’s ability with a pair of BC transducer to sense sound direction. In this paper, we discuss a way to minimize cross-talks in BC sound reproduction using a method called “cross-talk cancellation.” Ideally, the method requires transfer functions (TFs) from each of the BC transducers to each of the listener’s cochleae to accurately synthesize a cross-talk compensation (CTC) filter; however, a direct measurement of the TFs at the cochlea is not possible. Since the closest to the cochlea is the ear canal, we thus hypothesize that the TFs measured at each ear canal may be used to achieve cross-talk cancellation at the cochlea in the inner ear. Therefore we utilized a BC ear microphone to capture the vibration of the bony ear-canal caused by vibrating transducers on the mastoid. The filtered-x least mean square (FxLMS) algorithm was used to estimate the CTC filter. Experiments with and without cancellation were performed to determine the effective frequency range that could achieve cancellation in both the ear canal and inner ear perceived by participants.
	Primary ambient extraction for random sign Hilbert filtering decorrelation Lu Chen, Chuang Shi, Huiyong Li [Abstract] The primary ambient extraction (PAE) is a crucial step in spatial audio analysis and reproduction. The PAE decomposes a stereo mixture into separated primary and ambient components based on different spatial assumptions. The principal component analysis (PCA), although simple to calculate, often suffers from severe extraction error, particularly when the stereo mixture contains relatively strong ambient components. Reviewing the derivation of the PCA, it is found that the unbalanced ambient estimates incurred in the PCA degrades the extraction performance significantly. Therefore, this paper attempts to modify the classic PCA to enable the adjustable ambient components catering for various sound scenes. Comparing with the approximate solution to the ambient spectrum estimation, the proposed method is equivalently complex to implement and leads to higher extraction accuracy in selected sound scenes.
	Analysis of Head Rotation Trajectories During a Sound Localization Task Yukio Iwaya, Brian F. G. Katz [Abstract] Dynamic changes of the perceived Head-Related Transfer Function as a function of head movement have been shown to be an important cue in sound localization. To investigate the cognitive process of dynamic sound localization, quantification of the characteristics of head movement is needed. In this paper, trajectories of head rotation in a sound localization task were measured and analyzed. Listeners were asked to orient themselves towards the direction of active sound source via localization, being one of five loudspeakers located at 30∘ intervals in the horizontal plane. A 1 sec pink noise burst stimulus was emitted from different speakers in random order. The range of expected head rotations (EHR) for a given stimulus were, therefore, from 30∘ to 120∘. Head orientation was measured with a motion capture system (yaw, pitch, and roll). Analysis examined angular velocity, overshoot, and reaction time (RT). Results show that angular velocity increased as EHR increased. No relationship between overshoot and EHR was observed. RT was almost constant (about 260 ms) regardless of EHR. This may suggest that dynamic sound localization studies could be difficult for a short stimulus with duration less than 260 ms.
	Real time audio encoding/decoding system using MPEG-H 3D Audio toward advancement of terrestrial broadcasting technology Takehiro Sugimoto, Shuichi Aoki, Satoshi Oode, Tomomi Hasegawa, Hiroki Kubo, Hiroyuki Okubo [Abstract] A real time audio encoding/decoding system using MPEG-H 3D Audio (3DA) was developed. Research and development for advancement of terrestrial broadcasting technology has been carried out in Japan. The authors are engaged in improvement of audio scheme in broadcasting system, specifically on highly efficient audio coding, audio object handling by metadata, and rendering feature adaptive to preference of listeners and playback environment. Adopted coding technology is MPEG-H 3DA standardized in 2015. Coding efficiency of MPEG-H 3DA is twice as high as conventional AAC in 22.2 multichannel (22.2 ch) sound transmission. The developed system supports up to 24-channel beds with 4 objects and can render adaptively to typical playback environments, e.g., 22.2 ch, 7.1.4 ch, 5.1.2 ch, 5.1 surround and stereo. The encoder accepts and interprets audio objects handled with a serial representation of the Audio Definition Model and dialogue enhancement and dialogue replacement are realized using audio object. Transmission experiment with 8K video is also reported.
	Machine-learning-based estimation of reverberation time using room geometry for room effect rendering Ricardo Falcon Perez, Georg Götz, Ville Pulkki [Abstract] The reverberation of sound inside rooms has a significant impact on sound quality. In virtual reality, the reverberation of rooms should be reproduced with sufficiently high accuracy to produce plausible rendering of room acoustics. One of the most important parameters when rendering reverberation artificially is the reverberation time, and the value for it should be estimated from the geometry of the virtual room, for any given source/receiver position and orientation. While complete physical models are typically too computationally costly, an approximation can be made using simplified mathematical models such as Sabine of Eyring formulas, at the expense of reduced accuracy. This work shows a proof of concept for a machine learning based method that predicts the reverberation time of a room, using geometric information and absorption coefficients as inputs. The proposed model is trained and evaluated using a novel dataset composed of real-world acoustical measurements of a single room with 832 different configurations of furniture and absorptive materials, for multiple loudspeaker positions. The method achieves a prediction accuracy of approximately 90 % for most frequency bands. Furthermore, when comparing against the Sabine and Eyring methods, the proposed approach exhibits a much higher accuracy, especially at low frequencies.
	The Perceptual Effect of Reflective Surfaces on Acoustic Crosstalk Cancelation Using an 8-Channel Linear Loudspeaker Array Karim Bahri, Jens Ahrens [Abstract] We recently presented a user-tracked crosstalk cancelation system based on a linear array of loudspeakers. The system uses Ambiophonics in the low frequency range and a superdirective nearfield beamformer in the mid and higher frequency range. Efficient real-time execution is achieved by parameterization of pre-computed beamformer weights. The perceptual evaluation showed a considerable difference in the perception of binaural content depending on whether the system was operating in a free field or in a room with reflecting surfaces. In this contribution, we investigate the perceptual effect of different reflecting surfaces that are apparent in the reproduction room. Most notably, we separately investigate the effect of the floor reflection, which fuses with the direct sound from the loudspeakers, and the effect of the wall reflections, which arrive at later time instants.

Organization:

J. Li Q. Meng, Acoustics Lab, School of Physics, South China University of Technology

	Acoustically transparent sound presentation in hearing devices: algorithms, devices and models Simon Doclo, Henning Schepker, Florian Denk, Reinhild Roden, Matthias Blau, Birger Kollmeier [Abstract] Assistive hearing devices often have a low user acceptance due to a limited sound quality. Recently, the concept of acoustic transparency was proposed to increase the sound quality for hearing-impaired and normal-hearing users. Acoustic transparency of a hearing device refers to the acoustic perception of an open ear canal while the device is inserted in the ear. This can be achieved by applying an equalization filter to the output of the device such that the transmission properties of the open ear are obtained. In this contribution we give an overview of our approach to acoustically transparent sound presentation. We present recent advances in our custom earpiece design with multiple receivers and microphones as well as signal processing algorithms for robust acoustic feedback suppression and sound pressure equalization. We present results from subjective evaluations of a real-time demonstrator in terms of the overall quality compared to the open ear and outline future research directions, e.g., using individualized electro-acoustic models to alleviate the necessity of measuring the acoustic transfer functions within the device and to the eardrum. Funded by the Deutsche Forschungsgemeinschaft (DFG) - Projektnummer 352015383 - SFB 1330 A4 and C1.
	Cochlear-Implanted Children's Perception of Mandarin Tones in Normal Speech and Whispered Speech Xin Ding, Wentao Gu [Abstract] Since cochlear implants lack an efficient coding of fine structures of speech including F0, it is generally difficult for cochlear-implanted (CI) children to perceive linguistic tones, of which F0 is the primary acoustic cue. In whispered speech, however, F0 is missing and no longer serves as an acoustic cue of tones. Using a picture selection task, we compared the identification of Mandarin four tones in isolated syllables between two groups of 4-to-5-year-old children, i.e., CI and normal-hearing (NH) groups, in both normal speech and whispered speech. For normal speech, NH showed significantly higher identification than CI, regardless of tone which showed no significant effect. For whispered speech, both groups identified T3 better than other three tones, possibly due to their use of other cues for T3; Also, no significant difference was found between the two groups on all tones but T3, on which NH showed significantly higher identification than CI, suggesting that NH retrieved the voice quality cue for T3 better than CI. In both groups, identification rate was significantly lower for whispered speech than for normal speech, but the decrease in the rate was greater in NH than in CI, confirming that NH’s tone perception relied more on F0.
	How the temporal amplitude envelope of speech contributes to urgency perception Masashi Unoki, Miho Kawamura, Maori Kobayashi, Shunsuke Kidani, Masato Akagi [Abstract] Speech communicates non-linguistic and para-linguistic information as well as linguistic information. Our previous studies on noise-vocoded speech (NVS) showed that temporal modulation cues provided by the temporal amplitude envelope (TAE) affect how vocal emotion and speaker individuality are perceived. However, it is still unclear if temporal modulation cues affect the perception of urgency. Here, we experimentally investigated if the TAE of speech affects the perception of para-linguistic information, particularly urgency. We compared NVS in which the TAEs were identical to those of the original speech, and NVS in which the TAEs had undergone low-pass or high-pass filtering. Urgent scales were derived from a paired comparison of the results and used to investigate the relationship between the temporal modulation components and urgency perception. Our findings were: (1) the degree of urgency of the NVS stimuli was perceived as being similar to that of the original; (2) temporal modulation components of NVS upwards of 6 Hz were significant cues for urgency perception; (3) temporal modulation components of NVS downwards of 8 Hz were significant cues for urgency perception. The results suggest that temporal modulation cues in the TAE play an important role in urgency perception.
	Restoring Lost Speech Components with Generative Adversarial Networks for Speech Communications in Adverse Conditions Nengheng Zheng, Yupeng Shi, Yuyong Kang, Qinglin Meng [Abstract] Speech enhancement has been widely implemented to restore the speech quality for speech communication between humans or human and machine. For different speech communication scenarios and channel/environmental conditions, the types and degrees of speech distortion could vary significantly and many speech enhancement strategies have been developed accordingly. This study deal with a severe distortion problem, i.e., part of the spectral and/or temporal components of the speech are completely lost. The spectral loss is simulated by a transmission channel with very narrow passing bandwidth (lower than 2 kHz) which results in severely degraded speech quality; the temporal loss is simulated by packet loss up to 20% percent in massive communication which results in poor speech intelligibility. A generative adversarial networks (GAN) based speech enhancement scheme is proposed for restoring the missing spectral and temporal components with different network structure and parameters. A set of experiments have been conducted to evaluate the effectiveness of proposed enhancement scheme and promising results have achieved.
	Vocal Emotion Recognition in Mandarin-Speaking Cochlear Implanted Children Haitao Guan, Wentao Gu [Abstract] Previous studies have shown that cochlear-implanted (CI) children have big difficulty in vocal emotion recognition in comparison to the normal-hearing peers, mainly due to the lack of efficient coding of fundamental frequency (F0) in the CI devices, but few research has been conducted on preschool CI children, and even fewer on CI children speaking Mandarin, a tone language in which F0 conveys both lexical tone and emotion related intonation. The present study compared vocal emotion recognition between two groups of 4-to-5-year-old Mandarin-speaking children, i.e., CI and normal-hearing (NH) groups. Six semantically neutral sentences were recorded by a female speaker in three basic emotions, i.e., anger, joy, and sadness, among which anger and joy showed greater mean F0, F0 variability and F0 range than sadness. The children were asked to judge the type of emotion from the recorded utterances. For all three emotions, the CI group showed significantly lower recognition accuracy and longer reaction time than the NH group. The confusion mainly occurred between anger and joy in the NH group, while all three emotions were mutually confused in the CI group, confirming that Mandarin-speaking CI children were impaired in perceiving vocal emotions due to the lack of F0 processing ability.
	Intelligent background sound event detection and classification based on WOLA spectral analysis in hearing devices Feifan Lai, Kaibao Nie [Abstract] Audio signals of real-life hearing devices typically contain background noises. We aim to build a system model which can automatically separate background noise from noisy speech, and then classify background sound into predefined event categories. In this paper, we propose to use weighted overlap- add algorithm (WOLA) and deep neural networks(DNN) for sound event detection, from recordings of 11 realistic background noise environments(cafe, station, ...) combined with human speech at different SNR levels. In our approach, a sound trough detection algorithm is used to retrieve important background sound information. We use WOLA as an advanced algorithm which extracts spectral features by transforming a tiny fraction of time domain signal into frequency domain data represented in 22 channels. Moreover, a feed-forward neural network with one hidden layer and one output layer is used to efficiently recognize diverse event feature patterns, and then produce classification decisions based on confidence values. The overall detection accuracy has achieved 95%, while the event ëhallway’ has the lowest detection rate at 85%. This detection algorithm has the potential for improving noise reduction in hearing devices by applying distinct gain compensation in frequency channels for each particular detected event.
	Maximum Expanded Measurement Uncertainty: Hearing Aids Zemar Martins Defilippo Soares, Isabella Florêncio Cruz Da Silva, Nelson Mello Do Espírito-Santo [Abstract] For the study of measurement uncertainty estimation it is necessary to understand the model used to perform the measurements. In acoustic performance measurements of Hearing Aids the IEC 60118 part zero and part 7 present the method used for the measurements achievements. However, these standards present two possible acoustic environments where acoustic performance measurements of Hearing Aids can be achieved: Anechoic Chamber and Test Box. This work will estimate the maximum measurement uncertainty in acoustic performance measurements of Hearing Aids in Anechoic Chamber and Test Box. For the estimation of the maximum measurement uncertainty will be studied the main sources of uncertainties that lead to the greatest impact in the model used. A strong emphasis will be given to the source of uncertainty regarding the sound field where Hearing Aids are positioned for acoustic performance measurements. This investigation is based on results of experimental measurements. This work shows the differences found in the estimation of maximum measurement uncertainty in the Anechoic Chamber and in the Test Box.
	A pilot study of the relationship between mandarin chinese word and sentence recognition for the elderly Jianxin Peng, Jiazhong Zeng, Jiamin Liu [Abstract] In present study, the simulated binaural room impulse responses different reverberation time, were convoluted with the test word and sentence signals, and noise signal and reproduced by headphone. The subjective word and sentence recognition scores for elderly under different signal-to-noise ratio and reverberation time were obtained. The results show that the shorter the reverberation time, the higher the scores of word recognition and sentence recognition for the elderly. The higher the signal-to-noise ratio, the higher the scores of word recognition and sentence recognition. There is high correlation between the scores of word recognition and sentence recognition for the elderly.

Organization:

M. Melon, LAUM / Le Mans Université G. Behler, RWTH Aachen University

	Evaluation of Non-linear Distortion in Compression Chamber of a Low Frequency Horn Tomasz Nowak, Andrzej Dobrucki, Romuald Bolejko [Abstract] The subject of the paper is study of non-linear behaviour of air in horn-loaded low frequency speaker devices. Research and development of low frequency transducers lead to improvement their linearity of excursion trough linearization of suspension compliance and BL-factor as well as steadily increasing thermal power handling. The use of those latest transducers promises to achieve unprecedented efficiency and sound pressure levels, especially when combined with high compression horn- loading. The limiting factor in this case is nonlinearity of the air, which is particularly high in compression chamber. The volume of this chamber changes during operation of loudspeaker even by several dozen percent. This paper gives an insight on the influence of nonlinear compliance on nonlinear properties of loudspeaker through analyse and measurement of a prototype horn-loaded low- frequency loudspeaker device. As the driver an ultra- linear woofer has been applied. To evaluate the participation of the horn and compression chamber in total nonlinearity of the device, the measurement of the device with and without horn-loading have been done.
	Modelling of a Compression Driver using Lumped Elements Joerg Panzer [Abstract] Modelling linear response functions of a typical electro-dynamic compression driver with the help of lumped elements should be an easy task. However, it turns out that acoustic channels and slits inside such as device can go into strong coupling with the mechanical fundamental resonance. This paper demonstrates an extended electro-mechano-acoustical network and compares its response functions to measurement.
	Acoustic simulation for high intensity sound source with Helmholtz resonator Zinan Li, Bo Zhang, Qiqi Chen, Liheng Wang, Yutian Bai [Abstract] To study acoustic radiation characteristics of high intensity sound source used for the acoustic test of sound absorbing materials, a simulation model in the combination of horn-type loudspeaker and Helmholtz resonator, which are conceived for the design of a high intensity sound source in acoustic impedance tube, is developed by using software COMSOL. Fully coupled modeling related to multiple physical fields such as electromagnetic, mechanical, and acoustic ones is considered in this work, letting voltage be the input and sound pressure the output. Without the help of any simplified lumped parameter and only through the acoustic model constructed in COMSOL and accurate material parameters, the signal frequency response characteristics, output sound pressure, and electroacoustic transformation efficiency of the high intensity sound source can be obtained. The results show that the acoustic radiation characteristic of the high intensity sound source depends the interactions between horn-type loudspeaker and Helmholtz resonator, and its equivalent acoustic impedance characteristics is mainly dominated by Helmholtz resonator. It is also shown from results that the resonances of both horn-type loudspeaker and Helmholtz resonator may not only strengthen sound radiation of sound source, but enhance its electroacoustic transformation efficiency observably.
	Measurement of Loudspeaker Parameters: A Pedagogical Approach Antonin Novak [Abstract] Traditional ways of measuring linear parameters of an electrodynamic loudspeaker, i.e., fitting the measured signals of voltage, current, and velocity to a given model, are not very intuitive nor pedagogical. The method proposed in this paper is based on an educational approach that in the first step estimates the force factor and separates the electrical and mechanical impedance. In the second step, both impedances can be studied separately to highlight the importance of using models incorporating eddy currents on the electrical side and creep effect on the mechanical side.
	Effective Radiation Area (Sd) for Axisymmetric Loudspeakers Radiating in an Infinite Baffle Using a Near Field Analysis Luis Angelo Velarde Criado, Jorge N. Moreno [Abstract] In this paper, the effective radiation area (Sd) is obtained (theoretically and experimentally) for different loudspeakers. For this purpose, is considered a near field approach. This method is theoretically based in previous works that use the geometrical properties of different axisymmetric pistons and calculate the effective radiation area by positioning the pistons in an infinite baffle. In the case of loudspeakers, the calculations are redefined with the new geometrical conditions, in order to increase the precision of the results. Finally, these results are compared with the listed effective radiation area of the loudspeakers data sheets, and other techniques, in order to set the accuracy of the measurement.
	Measurement of the Back Loading of a Loudspeaker Mounted on a Closed Enclosure – A First Approach Jorge N. Moreno, Richard Rivera, Luis Angelo Velarde Criado [Abstract] In this paper, a simple way to measure the parameters related to the back loading of a loudspeaker mounted on a closed enclosure is presented. If the electro-mechano-acoustical parameters of the loudspeaker in free air are known or can be measure, then by measuring one more time the parameters of the loudspeaker mounted on a closed enclosure it is possible to calculate the back loading parameters due to the enclosure, including the filling material. This method could also be used to measure the effect of the filling material in the back cavity of midrange units, tweeters and compression drivers.
	Measurement of the Acoustic Impedance of Multi Resonant Systems - A Prior Step to Determine Parasitic Resonances in Magnetic Circuits of Loudspeakers Cesar Arthur Castro, Jorge N. Moreno [Abstract] This paper is focused on the measurement of the acoustic impedance of multi-coupled resonators using an impedance tube; results of measurements of different arrangements are shown in this work. It is well know that parasitic resonances can have negative effects on the frequency response of transducers, in order to analyze the effectiveness of the technique, geometries similar to the ones used in magnetic circuits of loudspeakers are also tested. The results are discussed in the conclusions.
	On the relevance of transducer measurements for real-world applications Martin Schneider [Abstract] Transducers are typically measured in laboratory environments like anechoic or reverberation chambers. The developer then can design his device to pre-determined demands. These are often linear frequency response, measured with plane or spherical waves in the free and/or diffuse field, and constant directivity. But demands may also diverge from these basic ideals, depending on the transducer’s intended application. In practice however, transducers are rarely used in such idealized environments. Transducers may have to be mounted inside a housing, held by clamps and stands, placed near reflective surfaces or bodies, and may even be hand-held. All these factors influence the transducer’s response. For the developer it is thus essential to measure the transducer in a variety of setups. These may include a set of near and far field measurements, with and without acoustically obstructive items, and may also bring up the necessity to simulate a human person and its body parts as acoustical objects. Integrating the results of such a set of measurements will hopefully confirm the transducers suitability for the intended application, but it may also lead to modified demands on the transducer design.
	Comparison of numerical methods for miniature loudspeaker modelling Juha Backman [Abstract] The small dimensions of the loudspeakers and earpieces found in portable audio devices pose challenges for the acoustical modelling. The millimetre scale dimensions imply that the sound propagation occurs to a significant extent within the viscous and thermal boundary layers, so viscothermal modelling is needed. The design question is further complicated by the geometric nonlinearity due to transducer diaphragm movement, and sound pressures inside the transducer that approach the levels that usually call for nonlinear acoustics models. The high viscosity forces can also lead to laminar vortex formation, which would then imply that the mo The paper studies an example of an axially symmetrical miniature transducer using different finite element approaches: - Small-displacement linear viscothermal - linear viscothermal with geometric nonlinearity due to the moving boundaries - nonlinear viscothermal with moving boundaries and nonlinear acoustics - compressible flow CFC with moving boundaries The results suggest that taking geometric nonlinearity into account is quite necessary for transducer modelling, while the typical nonlinear acoustics of pressure-dependent air compressibility are less significant. The CFD models indicate that vortex formation is a likely process in miniature transducer, but due to computational complexity the applicability of CFD models in everyday design is still limited
	Electrostatic All-Silicon MEMS Speakers for In-Ear Audio Applications - Acoustic Measurements and Modelling Approach Lutz Ehrig, Hermann Schenk, Franziska Wall, Anton Melnikov, Michael Stolz, Sergiu Langa, Bert Kaiser, Holger Conrad [Abstract] This paper presents the acoustic properties of a novel electroacoustic MEMS transducer. The transducer is based on electrostatic bending actuators with lateral deflection, i.e. in the chip plane. A large number of such bending actuators is arranged in pairs and move in phase opposition. By means of an appropriate sound guidance, areas of opposite phase pressure changes are separated, so that the acoustic short circuit is avoided and audible sound is generated by the system. In this paper, acoustic measurements recorded with a standard coupler are discussed and compared with the results of model calculations. The main focus is on system-specific features that provide important insights for the enhancement of future designs in terms of sound pressure level and distortion.
	Prediction of acoustic emission of a rigid electrodes DEAP loudspeaker Emiliano Rustighi, William Kaal, Sven Herold [Abstract] Dielectric Electro-Active Polymers (DEAPs) are lightweight materials whose dimensions change significantly when subjected to electric stimulation. One form of DEAP construction consists of a thin layer of dielectric sandwiched between two perforated rigid electrodes. They can be used as an actuator or a sensor and have the potential to be an effective replacement for many conventional transducers. This paper refers to their use as loudspeakers. To date, flat DEAP loudspeakers have been prototyped and tested but no numerical prediction of their acoustic performance has been presented. In this paper an elemental model is presented. The electro-dynamic behaviour of the electrodes and dielectric layers is taken into account. The acoustic impedance is calculated assuming baffled conditions. The impedances of the individual layers are stacked together and preliminary results are shown.
	Sensorless Measurement of the Acoustic Impedance of a Loudspeaker Romain Boulandet [Abstract] This paper presents an experimental approach for estimating the load acoustic impedance of a loudspeaker. The concept of impedance is commonly used to study the performance of a loudspeaker system. The acoustic load impedance is the effect of the acoustic field on the diaphragm that may affect the loudspeaker response, particularly in the low frequencies due to the coupling with the room modes. This impedance can hardly be modeled because it depends both on the physical characteristics of the loudspeaker and the complexity of the acoustic field in which it radiates. Conventional measurement techniques require external sensors such as microphones or accelerometers. The proposed approach relies mainly on the reciprocal mechanism of electrodynamic transduction, making it possible to use the loudspeaker as a sensor simultaneously. This work shows that the acoustic load impedance can be derived from an input electrical impedance measurement and the knowledge of the electromechanical parameters of the loudspeaker. An analytical model is given and data measured in actual rooms provides proof of concept. Potential applications are also discussed.
	Modelling of a MEMS transducer using approximate eigenfunctions of a square clamped plate Karina Abramova, Petr Honzík, Nicolas Joly, Stéphane Durand, Michel Bruneau [Abstract] The model of electroacoustic MEMS transducer with a moving square shaped clamped plate loaded by a thin fluid gap and a peripheral cavity is presented herein. The behaviour of the transducer, namely the thermal and viscous boundary layers effects originating in the fluid gap between the moving electrode and the fixed one and the strong coupling between the moving electrode displacement and the acoustic pressure field in the fluid gap, have to be described correctly by the model. The modelling approach proposed herein involving the integral method for describing the acoustic pressure in the fluid gap requires an analytical expression of eigenfunctions of the square shaped clamped plate. Such an approximate expression in form of two-dimensional cosine series proposed recently suffers from slow convergence and inaccurate boundary conditions. The solution proposed herein is based on the series expansion over the system of functions satisfying exactly the boundary conditions, which leads to faster convergence, hence lowering computational costs. The proposed eigenfunctions are described and the difference from the previous approximations is discussed. Finally, the acoustic pressure sensitivity of the transducer is presented and compared to the results of a reference finite element model.
	Temperature effects on the mechanical-acoustic properties of condenser microphones: experimental characterization Cécile Guianvarc'H, Thomas Lavergne, Lara Risegari, Petr Honzík, Dominique Rodrigues, Roberto M. Gavioso [Abstract] Condenser microphones are originally designed for airborne acoustic measurements. For this reason, their electro-mechanical behaviour is well-known and characterized almost in atmospheric conditions. However, more and more applications, such as acoustic thermometry, gas metrology or thermoacoustics, require the use of electrostatic transducers (as acoustic transmitters or receivers) in gas conditions that differ significantly from atmospheric ones. Previous experiments using the electrostatic actuator technique evidenced the relevant influence of the static pressure and the type of gas on the frequency response of condenser microphones. The same technique is used here to characterize the mechanical properties of a 1/4” microphone’s membrane in vacuum at temperatures between 80 K and 300 K. The resonance properties of the membrane are then obtained and expressed as function of the temperature. Thus, they are used in comprehensive models of condenser microphones whose validity as function of the temperature is experimentally checked. Also, an improved model of condenser microphone coupled to an electrostatic actuator is developed for the sake of this work. By improving the supporting models and experimental methods, this work is expected to drive the design of new electrostatic transducers and promote an advance of the calibration procedures, as required for specific applications.
	Analytical modelling of a MEMS transducer composed of a rigid micro-beam attached at one end to a flat spring moving against a reduced-size backplate Petr Honzík, Antonin Novak, Stéphane Durand, Nicolas Joly, Michel Bruneau [Abstract] The use of planar micro-beams as moving parts of acoustic and electroacoustic devices has increased recently because of their geometrical simplicity, hence lowering fabrication costs. The precise modelling of such devices is then of interest. The miniaturized transducer proposed herein is composed of a planar rigid micro-beam attached at one end to a flat spring (the other end remaining free) surrounded by thin slits and loaded by a thin fluid layer (situated in the gap between the micro-beam and a reduced-size backplate) and a small cavity, both being placed behind the beam. Such a configuration reduces the overall size of the device (no need of an external cavity) and enables to adjust more parameters comparing to the case of the backplate of the same size as the one of the micro-beam. The thermoviscous damping effects originating in the fluid-filled parts of the device (slits, air-gap, and cavity) are taken into account. As a result of the model, the displacement of the micro-beam is calculated and compared with the reference finite element solution, the acoustic pressure sensitivity of the transducer is finally presented and discussed.

Organization:

W. De Roeck, KU Leuven W. Desmet, KU Leuven/Member of DMMS Lab, Flanders Make

	Generalized Method of Describing Acoustic Duct-Like System as a Multiport Anna Snakowska, Jerzy Jurkiewicz [Abstract] The paper presents general method to analyse acoustic mufflers or other duct-like systems assuming propagation of a multimode wave and applying the multiport method. Apart from the main propagating tube and expansion chambers the analysed device may contain segments of annular duct (concentric or not with the central tube) branch resonators, cavities and cavity diaphragms. Systems of such complicated geometrical structure have been widely analysed under the assumption that the wave propagating through any element is the plane wave. Thus the problem might have been analysed by means of the two-dimensional transfer matrices (or others equivalents) derived from conditions of conservation of the acoustic pressure and the volume velocity across each discontinuity. However, the assumption of the pure plane propagation across every element of the acoustic system imposes strong limitations on the excitation frequency or the radius of the duct segments as the Helmholtz number should not exceed the value 1.93 which represents the non-dimensional cut-on frequency of the first circumferential mode. If the excitation and the system are axially symmetric the limited value of the Helmholtz number raises to 3.83. The analysis is carried out by means of the scattering matrix which is finally transformed into transmit matrix.
	Characterization of the centrifugal fan noise generated in residential HVAC systems using in-duct measurements Sarah Van Tricht, Pieterjan Broos, Hervé Denayer, Maarten Vanierschot, Wim De Roeck [Abstract] The acoustic characterization of a centrifugal fan in a residential HVAC system is carried out through the determination of the total source spectrum of the sound generated by the fan under varying operating conditions. The most commonly used characterization technique, described by ISO 5136, requires an anechoic termination and the possibility to insert a microphone in the duct. Since this approach largely depends on the efficiency of the anechoic termination, an active one-port characterization using flush mounted microphones is adopted in this research. One of the major advantages of this approach is the fact that the characterization is carried out independent of downstream reflections, thus not requiring the construction of an anechoic termination. At first, the passive property of the fan, i.e. the reflection coefficient of the propagating waves at the inlet of the fan, is determined. Subsequently, the downstream reflection component of the test rig is characterized. Finally, both passive models are combined to compute the active source spectrum describing all noise generation and dissipation phenomena of the fan. The acoustic characterization is compared to existing scaling laws, such as described e.g. in VDI 2081.
	Trained Algorithms for Mode Decomposition in Ducts Stefan Sack, Mats Åbom [Abstract] Acoustic waves in duct systems become non-linear if their pressure amplitudes or particle velocities exceed critical levels. The critical levels depend on the pressure and velocity of the ambient fluid and the geometry of the flow channels. A microperforated plate mounted inside a duct, for example, may have a very large local particle velocity inside the perforations even for relatively low acoustic excitation levels. Classical models for in-duct sound such as the acoustic two-port are only valid within the linear region and, therefore, under certain circumstances describe the scattering through perforations inadequately. Although some attempts have been made to incorporate non-linearity into existing methods, experimental studies still suffer from varying particle velocities depending on the frequency. We developed an extension to the classical two-load method that uses a combination of excitation sources to guarantee a constant particle velocity for all tested frequencies. By means of this method, a non-linear scattering matrix can be extracted. As a test case, the particle velocity-dependent scattering of a perforated panel inside a circular duct is presented.
	Prediction of the nonlinear response of an acoustically forced flow with linearized Navier-Stokes equations Abel Faure, Claire Bourquard, Nicolas Noiray [Abstract] The prediction of self-sustained aeroacoustic oscillations has attracted significant research interest for decades. However, quantitative predictions of whistling amplitude on practical configurations are still challenging. In this study we use particle image velocimetry to record flow data at an acoustically excited T-Junction subject to low-Mach grazing flow. Data is recorded for several acoustic excitation amplitudes and frequencies. This allows to formulate a model predicting the non-linear response of the system with Navier-Stokes equations linearized around different mean flows, which were measured with increasing excitation amplitudes. This linearized perturbation approach gives a good prediction of the spatial distribution of the velocity fluctuations and predicts their amplitude to a reasonable degree. The study concludes with a discussion on the benefits and drawbacks of modeling the aeroacoustic coupling at the shear layer with the incompressible Navier-Stokes equations linearized around the mean flow.
	Self-Excited Vibration of Valve Caused by Hydrodynamic Force Xiangdong Liang [Abstract] Liquid vibration is the source of duct noise and is main reason of pipe connector to be loosen and leaking out working medium from the connector also. Besides machine linked to the pipe line, valve in the pipe line is also one kind of vibration source to the pipe line because liquid passing through the valve can cause valve to vibrate and radiate noise. Thermostat control valve is a kind of self-adjust valve used in oil pipe line for protecting machines. To control the oil temperature in selected range the thermostat control valve can adjust the oil flows to the heating branch pipe and cooling branch pipe as the oil temperature is changed. But when it adjusts flow of oil, vibration and noise will be taken place in the valve. And this kind of liquid vibration may damage the thermal rod of the thermostat control valve. The mechanism of how this kind of valve liquid vibration taken place will be shown in this paper. Some of suggestion about decreasing or avoiding this kind of vibration on the oil pipe line will be given in this paper based on the conclusion of investigation.

Organization:

N. Henrich, Université Grenoble Alpes M. Döllinger, Division of Phoniatrics and Pediatric Audiology, University Hospital Erlangen

	Acoustic Source Term Interpolation in Hybrid Aeroacoustic Simulation of Human Phonation Michael Weitz, Stefan Schoder, Paul Maurerlehner, Sebastian Falk, Michael Döllinger, Manfred Kaltenbacher [Abstract] We use a hybrid aeroacoustic approach that relies on an incompressible flow simulation resolving relevant turbulent scales. Based on the flow results, we compute the acoustic source terms on the flow grid and perform a conservative interpolation to the acoustic grid, on which we solve the perturbed convective wave equation to obtain the acoustic field. This conservative transformation of the acoustic sources from the flow grid to the acoustic grid is a crucial step to allow coarse acoustic grids without reducing accuracy. An advanced cut-volume-cell approach, guaranteeing high accuracy both for regions where the flow grid is finer and for regions where it is coarser than the acoustic grid, is employed. In this framework, the radial basis function method is incorporated to compute spatial derivatives of the flow data as necessary in the computation of the acoustic sources. The simulations demonstrate the applicability of the cut- volume-cell interpolation method to obtain an accurate transfer of the acoustic sources from the flow grid to the acoustic grid. The acoustic results changed very slightly even for a reduction of elements in the source region of more than 10 times compared to the initial acoustic grid, leading to a strong reduction in CPU-time.
	Acoustic characterization of an averaged vocal tract model based on the MRI data of professional tenors Judith Probst, Alexander Lodermeyer, Sahar Fattoum, Matthias Echternach, Stefan Becker, Stefan Kniesburges [Abstract] Human speech is created in the larynx and the vocal tract. For vowels, the basic sound is generated by the oscillating vocal folds and filtered by the vocal tract to produce the typical formants. In the present study, we investigate the influence of the vocal tract on phonation. Based on Magnet Resonance Imaging (MRI) data, the acoustic behavior of the vocal tract can be characterized using numerical and experimental models. However, previous studies only analyzed single vocal tracts that were directly obtained from individuals. To investigate the basic influence of the vocal tract on phonation, a vocal tract model is necessary that shows preferably little individual but the typical acoustic characteristics of the specific vowel. Hence, we computed an average vocal tract geometry based on six simplified individual vocal tracts that were extracted from MRI images of professional tenors. The resulting vocal tract model was evaluated in an experimental setup including artificial vocal folds. The comparison with the tenors’ vocal recordings shows that the model reproduces the acoustic properties well. Hence, the average vocal tract model presented here is suitable for future investigations of the influence of the vocal tract on phonation.
	Influence of the vocal tract on voice directivity Rémi Blandin, Manuel Brandner [Abstract] Voice directivity induces variations of the sound amplitude and frequency content with the direction. Voice directivity is important for the efficient transmission of speech and singing. Therefore, it is taken into account both by the artists and the designers of concert halls in order to enhance the quality of artistic performances. While the shape of the head and the torso has long been considered to study this phenomenon, the influence of vocal tract shape has received very little attention. However, it has recently been shown that it can influence voice directivity through the internal acoustic field and the frequency content. The contribution of the vocal tract to the voice directivity is characterized through physical modelling and direct observation of speakers and singers. In particular, the role of higher order propagation modes is discussed as well as strategies to control directivity through the frequency content of voice.
	Finite Element Simulation of /asa/ in a Three-Dimensional Vocal Tract using a Simplified Aeroacoustic Source Model Marc Arnela, Oriol Guasch [Abstract] The numerical simulation of fricative sounds in three-dimensional (3D) vocal tracts typically involves hybrid Computational Aeroacoustics (CAA) approaches. Unfortunately, those are very costly and require from super-computer facilities to produce a mere few milliseconds of sound. The problem becomes prohibitive if one not only aims at generating a single fricative, but also sequences containing both, vowels and fricatives. It then seems wise to try to approximate somehow the flow noise sources, so that only an acoustic simulation becomes necessary avoiding the very demanding computational fluid dynamics step in CAA. This can be done to different levels of precision. In this work, it is suggested to follow a similar methodology to that in one-dimensional techniques, but applied to 3D dynamic vocal tracts. Vowel sounds are generated introducing glottal pulses at the vocal tract entrance (glottis), while a monopole source consisting of filtered white noise is activated in a location immediately downstream of the constriction where a fricative sound is articulated. Acoustic wave propagation in a 3D dynamic vocal tract is simulated using a stabilized Finite Element Method (FEM) for the wave equation in mixed form, set in an arbitrary Lagrangian-Eulerian (ALE) framework. The sequence /asa/ is produced as an example.
	Glottal Opening Measurements in VCV and VCCV Sequences Yves Laprie, Benjamin Elie, Angélique Amelot, Shinji Maeda [Abstract] Many studies on speech acoustics and production use articulatory synthesis as a framework to investigate the relationship between articulatory gestures and acoustic features. Although many supraglottal articulatory models are available, usually built from vocal tract imaging acquisitions, glottal gestures are commonly modeled with simple geometric primitives which do not necessarily reflect reality. This is an importa issue as glottal opening gestures are very important to control the glottal flow at the vocal tract input and the supraglottic acoustic flow. This study is a first step towards the development of a database of realistic glottal gestures which will be used to design the glottal opening dynamics in articulatory synthesis paradigms. In the first part of this paper, we present experimental measurements of glottal opening dynamics in VCV and VCCV sequences uttered by real subjects, thanks to a specifically designed external photoglottographic device (ePGG). The corpus was designed to highlight the differences in glottis opening between fricatives and stops. The existence of different patterns of glottal opening is evidenced according to the class of the consonants. A numerical study is then used to show the influence of these patterns on the production of sounds and on the coarticulation.
	Formation of stationary intraglottal vortices during phonation Liran Oren, Sid Khosla, Charles Farbos De Luzan, Alexandra Maddox, Ephraim Gutmark [Abstract] According to the myoelastic-aerodynamic theory of voice production, the vibrations of the vocal folds occur due to interaction between the fold’s elasticity and intraglottal aerodynamics. The shape of the glottal opening changes from converging to diverging during the opening and closing phases of the folds vibration, respectively. The divergent shape causes the glottal jet to separate from the medial wall and rotational flow develops in the regions between the wall and the jet. The formation of these intraglottal flow separation vortices (FSV) was shown both experimental and computationally, but their role in the phonation process is still an ongoing debate. Few models for the vibration mechanism have predicted that FSV would form as shedding vortices, thus convecting downstream with the glottal jet. In the current study, we used time-resolved phase-locked PIV measurements of intraglottal flow in a canine larynx model. Results show that FSV develop and remain as recirculation zone near the superior aspect of the folds. It is important to correctly model FSV because they can produce negative pressure near the superior aspect of the folds, which we hypothesize can increase maximum flow declination rate (MFDR). MFDR is important because it is clinically related to loudness and vocal efficiency.
	Modeling the pre-phonatory vocal fold posture in the larynx model SynthVOICE Stefan Kniesburges, Reinhard Veltrup, Sahar Fattoum, Anne Schützenberger [Abstract] The basic aim of synthetic larynx models is to reproduce the phonation process. Thus, we develop the synthetic larynx model SynthVOICE that incorporates the control of the pre-phonatory posture (ad-/abduction) and the elongation of the vocal folds (VFs). The aim is to reproduce the physiological and pathological phonation process especially dysfunctions of the muscular and nerval control such as muscle tension dysphonia. The model is manufactured in a multistage pouring process using liquid silicone mixtures with a different stiffness. It is composed of a cylinder mimicking the thyroid cartilage, the VFs and five manipulator devices to control the pre-phonatory VF posture and tension. Both, single- and three-layer VF models can be inserted that are based on the M5 shape. The pre-phonatory settings are controlled by a mechanical setup that exhibits seven linear and two rotational traverse tables. A Matlab script determines the appropriate displacements and angles of the traverse tables to move the VFs in a specific posture. To validate the SynthVOICE model, we present measurements of the oscillation threshold pressure, the oscillation frequency and the produced acoustic signal for different settings of VFs posture. The Else-Kröner-Fresenius Stiftung is highly acknowledged for its funding (grant-no. 2016_A78).

Organization:

S. Moreau, University of Sherbrooke S. Becker, Institute of Process Machinery and Systems Engineering, University of Erlangen

	Analysis of Sound Source Localization in an Axial Fan Seyed Mohsen Alavi Moghadam, Matthias Meinke, Wolfgang Schröder [Abstract] Numerical analyses of the tip vortex system of a ducted axial fan are conducted by a hybrid fluid-dynamics-acoustics method. First, large-eddy simulations are performed to investigate the dynamics of tip clearance flow for various tip-gap sizes and to determine the acoustic sources. The simulations are performed for a single blade out of five blades with periodic boundary conditions in the circumferential direction on a multi-block structured mesh with 140$\times10^6$ grid points. The turbulent flow is simulated at a Reynolds number of 9.36$\times10^5$ at undisturbed inflow condition and the results∼\cite{Moghadam2019} are compared with experimental data∼\cite{Tao2013}. The results show that increasing the tip-gap size results in various vortices in the tip-gap region, i.e., tip-leakage, separation, and induced vortices, which enlarge the diameter and the strength of the main tip vortex. For the largest tip clearance, spiral vortex breakdown occurs at the design operating condition. Second, the acoustic field on the near field is determined by solving the acoustic perturbation equations (APE)∼\cite{Ewert2003} on a mesh for a single blade consisting of approx. 1060$\times10^{6}$ grid points. The results show that the larger the tip gap size the higher the broadband noise level∼\cite{MoghadamACTA2019}. Detailed results will be presented at the conference.
	Prediction Methodology of Broadband Noise from a Cooling Fan Soichi Sasaki [Abstract] The methodology for prediction on broadband noise of a cooling fans using in HVAC system will be proposed. Because such the fan cools the heating device by its own wind velocity, the fan has no duct and also there is little static pressure rise. This methodology has an originality applying a modified blade element momentum theory for the prediction of the broadband noise. The slip factor of the blade element is evaluated based on the measured aerodynamic characteristics and the Euler pressure. The slip factor can give the attack angle and relative velocity of the blade elements in the design phase. The characteristics of attack angle of the broadband noise generated from the isolated blade are measured by wind tunnel test. The broadband noise generated from the cooling fan is predicted by superimposing the two dimensional blade elements in the span direction. In the full length paper, the procedure of the prediction methodology of the broadband noise, the result of wind tunnel test of the isolated blade and the comparison of the actual measurement noise spectra generated from the fan and its predicted value will be presented.
	Numerical Investigation of Hydrodynamic/Acoustic Splitting Methods in Finite Volumes including Rotating Domains Joscha Piepiorka, Otto Von Estorff [Abstract] In today’s product development, the behavior of fans is increasingly important. Especially the noise emission is a major challenge to be handled by the manufacturers. Alongside expensive experimental investigations, reliable numerical simulations to predict the flow induced noise are gaining more and more importance. The paper will present first numerical results of an axial fan benchmark case [1]. A simplified model of the experimental setup is used for a scale-adaptive CFD simulation. Based on incompressible CFD results, the acoustic quantities are modelled by a finite volume approach, which has been developed in the software framework OpenFOAM. Different hydrodynamic/acoustic splitting methods are used to investigate the flow induced noise including the rotating domain of the fan. Special focus is placed on velocity terms and their influence onto aeroacoustic source and wave propagation effects. For this purpose, different hydrodynamic/acoustic splitting methods are introduced and reformulated. The resulting sound fields are computed and interpreted based on the theoretical assumptions. Finally, the computational effort in the finite volume framework is discussed and the results are compared to the experimental data [1: Zenger, Florian, et al. A benchmark case for aerodynamics and aeroacoustics of a low pressure axial fan. No. 2016-01-1805. SAE Technical Paper, 2016].
	Vortex sound based calculations for the aeroacoustic noise of a centrifugal fan Hakan Dogan, Martin Ochmann, Chris Eisenmenger, Stefan Frank [Abstract] Radial ventilators are widely used in industrial applications such as home appliances, automobile industry and air-conditioning devices. In this work, we investigate the aerodynamic and aeroacoustic performance of a radial ventilator with nine backward curved blades by means of computational fluid dynamics and aeroacoustics. In a few previous publications by the authors, some experimental and numerical results were presented and compared. Regarding the aeroacoustic computations, the far field noise was calculated by solving the Helmholtz equation using the boundary element method (BEM) or the finite element method (FEM). Moreover, FEM calculations using Lighthill’s analogy (Lighthill’s wave equation) were also done. In the current paper, we perform vortex sound theory based aeroacoustic calculations, namely using Powell’s analogy. The results obtained using vortex sound theory will be compared with the theoretical/numerical approaches mentioned above, and with the experimental results. As stated by some previous researchers, although Lighthill’s and Powell’s formulations pose only slight differences mathematically, they exhibit considerably different numerical robustness. The application, numerical stability and accuracy of the two methods with regards to the current problem will be investigated and compared in this paper.
	Experimental Investigation of the Influence of different Leading Edge Modifications on the Sound Emission of Axial Fans downstream of a Heat Exchanger Felix Czwielong, Florian Krömer, Chaitanya Paruchuri, Stefan Becker [Abstract] Axial fans have a major impact on the sound emission of technical systems. One of the common applications of low-pressure axial fans is their usage in air conditioning and cooling circuits. In most cases the fan operates downstream of a heat exchanger in this system. The suction side heat exchanger changes the upstream flow field of the fan and consequently the fan acoustic behavior compared to undisturbed inlet conditions. Investigations on sound emission of airfoils showed that leading-edge modifications like serrations or slits reduce the turbulence-airfoil interaction noise. In this study, the impact of three different leading-edge modifications (straight leading edge, sinusoidal serration and slits) on the sound emissions of a forward skewed low-pressure axial fan was examined. The inflow conditions varied between a free inflow and a disturbed inflow, generated by a heat exchanger located on the suction side of the axial fan. The results show that the modification of the leading edge has a positive influence on the sound radiation of the axial fan for both inlet conditions discussed above.
	Aeroacoustic Simulation and Experimental Validation of Sound Emission of an Axial Fan Applied in a Heat Pump Andreas Lucius, Marc Schneider, Stefan Schweitzer-De Bortoli, Tom Gerhard, Thomas Geyer [Abstract] The reduction of noise emission of heat pumps is a major development target for the appliance manufacturer. An important source of sound is the fan. The fan is operating at disturbed inflow conditions due to installation in the appliance. Additionally small scale turbulence generated by the heat exchanger further increases fan noise emission. Noise emission may be reduced via improved installation or specific fan design. The aim of the project was to measure flow and turbulence characteristics inside the heat pump via constant temperature anemometry as well as noise emission. The acquired data set was then used for validation of CFD and CAA simulations. The effect of inflow turbulence was analyzed for configurations with and without heat exchanger. Hot wire measurements show increased average turbulence levels of 20 percent downstream the heat exchanger. In a second step URANS simulations were conducted for both configurations. The heat exchanger was modelled as porous medium. Turbulence properties could be matched with experiments with reasonable boundary conditions at the heat exchangers outlet. A LES-based CAA simulation of the heat pump was done for the configuration without heat exchanger. There is some deviation in flow rate but quite good prediction of sound power.
	Experimental investigation of the noise reduction of a plug fan with leading-edge serrations Ignacio Zurbano-Fernandez, Alain Guedel, Mirela Robitu [Abstract] In previous years, leading-edge serrations have been successfully used as a noise-reduction technique applied to airfoils. More recently, this technique has been also used and validated on axial fans. In this study, sinusoidal leading-edge serrations have been integrated on a plug fan (centrifugal fan without volute). The outcome of RANS simulations of the non-serrated baseline impeller has been used to define an assumed optimum serration geometry in terms of noise reduction. This optimum geometry has been used as a reference to manufacture three impeller prototypes of various geometries, defined by their amplitude (2h) and wavelength (λ). This paper will present the results of the test of the prototypes in a reverberant room, where noise and air performance measurements are done simultaneously. For each impeller, five fan operating points will be presented, and for each one of the points the sound power level is measured at both the inlet and the outlet of the fan. This allows us to assess the effectiveness of serrations at different operating points, while checking the impact on the fan efficiency. The results will also be compared with the non-serrated baseline fan.
	Computational Aeroacoustics of an Axial Fan with Leading Edge Serrations Stefan Schoder, Florian Krömer, Michael Weitz, Manfred Kaltenbacher, Stefan Becker [Abstract] In recent years a trend towards modifications of the leading edge geometry in airfoils for reducing the sound emission and improving the stall behavior has been observed. However, low-pressure axial fan design methodologies for a decreased sound radiation almost rely exclusively on the use of fan blade skew rather than on leading edge modifications so far. Hence, with this study we aimed to investigate the sound reduction capabilities of leading edge serrations applied to low-pressure axial fans using the efficient perturbed convective wave equation. For this, we used a reference fan with straight leading edges and a fan with serrated leading edges that has been studied in experimental investigations. The comparison of the CFD and CAA simulations with experimental results show good agreement. Furthermore, we analyzed the sound reduction mechanisms of leading edge serrations based on the CAA results. In general a sound reduction was obtained with leading edge serrations without influencing the fan operating point. The study demonstrates the applicability of this aeroacoustic approach for axial fans with modified leading edges.
	Axial Ventilator Tip Gap Noise Prediction from Discontinuous Galerkin based CAA with Stochastic Vortex Sound Sources Lev Liberson, Markus Lummer, Michael Mößner, Roland Ewert, Jan Werner Delfs [Abstract] A prominent noise source of axial ventilators emerges from the complex flow within the tip clearance between fan blades and surrounding encasement. To obtain an aeroacoustic fan design, an associated process must be acceptably efficient while providing a sufficiently amount of physical modeling depth. The mechanisms of noise generation as well as the sound propagation in highly inhomogeneous flow fields have to be captured. A promising method is developed at the DLR by using the in-house CAA propagation solver DISCO++ coupled with the stochastic acoustic source generator Fast Rand Particle Mesh Method (FRPM). Utilizing the Discontinuous Galerkin method the Acoustic Perturbation Equations (APE) are solved on an unstructured tetrahedral mesh, while the acoustic sources are reconstructed on a Cartesian background grid. The prediction capability of the method is evaluated by computing the noise generated by a ducted axial ventilator at two different tip clearance variations. While the trailing edge noise is well captured, the tip gap noise increase is not properly predicted in terms of spectral shape - presumably due to anisotropic length scale influence not properly recognized by the current stochastic model. First simulations based on an enhanced anisotropic length scale model will be presented and results discussed.
	Aeroacoustic evaluation of the forward-curved fan inlet flow Jurij Gostiša, Tadej Novaković, Jurij Prezelj, Marko Hočevar [Abstract] Forward-curved fans have been extensively used in various industrial and residential HVAC applications. Regarding their favourable characteristic properties, low noise emissions and small size, forward-curved fans have found use in various applications with requirement for high flow rates at moderate pressures and efficiency not being of primary importance. Design of the fan’s inlet flow channel is heavily influenced by limited space available, which results in adverse flow aerodynamics. One of the main reasons for poor efficiency of the forward-curved fan is the flow separation, which starts at the inlet, develops into the blade area and then into the volute. Such vortical flow occupies about a third of the rotor’s width, resulting in a significant flow rate and efficiency reduction. Local measurements of flow velocity using hot wire anemometry were performed and a novelty approach with psychoacoustic metrics adopted to velocity signal processing was used to improve the understanding of the aerodynamic phenomena in the inlet area. The findings exhibit distinct areas and properties of adverse vortical flow and are used in the further inlet channel design process.
	Influence of the Mesh Size on the Aerodynamic and Aeroacoustics of a Centrifugal Fan using the Lattice-Boltzmann Method Rebecca Schaefer, Martin Boehle [Abstract] Due to market requirements and ever stricter guidelines, noise emissions of fans are becoming increasingly important in addition to energy efficiency. For this reason, developer aims not only at more efficient running fans but also ever quieter fans. In order to keep the development process cost-effective and short, numerical investigations are used today. The Lattice-Boltzmann method (LBM) is a relatively new method promising to calculate aerodynamics coupled with aeroacoustics. The present paper deals with the simulation of the aerodynamics and aeroacoustics for a centrifugal fan using the commercial LBM Code PowerFLOW of the Exa Corporation. The focus is on the variation of the mesh size. The aim is to determine the smallest cell size and the geometry defined for refinement in such a way that both the flow and acoustics of the centrifugal fan can be accurately predicted with an acceptable total computing time. At the end a total of four operating points are analysed. The results of the LBM simulation are compared and validated with measurement results. Flow information in terms of the pressure rise as a function of the volume flow rate is well captured. In addition, the noise is well predicted with respect to blade passing frequency.
	Research on Aerodynamic Noise Calculation and Noise Reduction Design of Multi-blade Centrifugal Fan Jin An Huang, Yang Xiang, Chaojun Jiang [Abstract] Taking the multi-blade centrifugal fan in an air- conditioning system as the object, the three-dimensional modeling of the internal fluid of the fan is carried out. The steady-state and non-steady-state calculations are carried out in the CFD software, obtaining the internal flow field information, with calculation result imported into Noise estimation which was performed in the LMS Virtual.lab. Besides, the accuracy of the fan aerodynamic noise calculation was verified against the experimental results. Through the analysis of the mechanism of the aerodynamic noise of the fan, the main noise source of the aerodynamic noise of the fan is located at the impeller and is closely related to its internal flow field distribution and its structure. For the fan products that have been put into production, it is difficult to modify the main structure including the volute, the impeller, the blade size, etc., and only local structure can be redesigned to achieve the purpose of noise reduction. Therefore, the blade perforation design is adopted. By setting reasonable perforation parameters, the eddy current shedding around the blade can be reduced without changing its performance, so as to reduce the pressure pulsation of the blade surface and the aerodynamic noise of the fan.

Organization:

P. Croaker, UNSW Australia M. Liefvendahl, Swedish Defence Research Agency (FOI)

Simulation of flow noise generated by the interaction of inflow turbulence with the leading edge of a foil Mattias Liefvendahl [Abstract] Noise generated by the interaction of inflow turbulence with a lifting surface may be of interest for a number of hydrodynamic applications, including a propeller operating in non-cavitating condition. In the present work, simulation methods are applied for the prediction of the radiated noise. The experimental set-up of Santana et al. (2016) will be investigated at a chord- based Reynolds number, Re = 256 000, and Mach number of 0.09, with strong homogeneous inflow turbulence interacting with the foil. The large-eddy simulation (LES) technique will be employed to predict the flow, with synthetic turbulence fluctuations applied at the inflow boundary. The LES will be employed to compute the acoustic source terms of an acoustic analogy (Liefvendahl and Bensow, 2018). The spectrum of turbulence upstream of the airfoil leading edge will also be extracted from the LES and combined with semi-analytical scattering models for leading edge noise. The flow field and the radiated noise predicted with the LES and semi-analytical techniques will be compared with experimental data from the literature. Liefvendahl & Bensow, 32nd Symp. on Naval Hydrodynamics, 2018. Santana et al., J. Sound and Vibration, 383:349-363, 2016.

Noise from a blunt edged flat plate in a reverberant water tunnel Paul Croaker, James Venning, Mahmoud Karimi, Paul Brandner, Con Doolan, Nicole Kessissoglou [Abstract] Large eddy simulations of flow past a blunt edged flat plate located in a reverberant water tunnel are presented. The flow is at a Reynolds number based on chord of 6.8e6 and a Mach number of 5.3e-3. Hydrodynamic pressures on the surface of the plate are extracted and processed to investigate the wavenumber- frequency content of the pressure fluctuations. Unsteady velocities at the mid-span plane of the plate are also processed to extract volumetric flow-induced noise sources in the vicinity of the blunt edge. The surface pressures and volumetric source terms are then combined with semi-analytical models of trailing edge scattering to predict the pressure fluctuations on the wall of the water tunnel. These analytical models were originally developed to analyse far-field sound produced by flow past a structure in air. This work investigates the suitability of applying these scattering models to near-field prediction of flow-induced noise in a reverberant water environment. The results are compared with experimental measurements taken in the Cavitation Research Laboratory at the Australian Maritime College, University of Tasmania.

Organization:

T.P. Chong, Brunel University London

	Aerofoil trailing edge self-noise reduction by Surface Mounted Attenuation Devices Edvard Schroeder, Tze Pei Chong, M. Kamruzzaman, Jeremy Hurault, Phillip Joseph [Abstract] This paper presents the results of a preliminary experimental study into the effect of add-on type finlets on the aeroacoustic performance of a wind turbine aerofoil. The content mainly deals with the turbulent boundary layer -trailing edge broadband noise characteristics subjected to the add-on finlets. The test program seeks to test various combinations of finlet height and spacing present on the (1) suction side only, (2) pressure side only, and (3) both suction and pressure sides. Each finlet parameter configuration is tested at jet velocity U=30m/s and 45m/s, at geometric angles of attack AOA (geometric) = 10∘, 0∘, -10∘. Based on the data acquired thus far it is observed that a finlet usually performs best with increasing height and decreasing spacing. Under configurations (1) and (2), finlets are able to produce up to -2dB Sound Power Level broadband noise reductions, where configuration (3) offers broadband reductions of up to -7dB Sound Power Level, with potential to observe more reductions once optimised. The main parameter for improved performance of finlets is the spacing rather than the height. Finlet height does not offer significant impact on the performance if the spacing is not optimal.
	The Effect of flow-permeable material on the Flow Field and the Aerodynamic Noise of Cylinders Thomas Geyer, Richard Peschel [Abstract] The flow around circular cylinders is one of the major aeroacoustic noise source mechanisms. Such a cylinder represents a simple model for technical applications like parts of the landing gear of airplanes, train pantographs, antennas or other protruding parts of vehicles. A reduction of the flow noise is especially desirable since it contains both a broadband component and tonal components due to the regular vortex shedding. One approach to reduce this noise is to cover the cylinder with flow-permeable materials, such as foams or porous rubbers. In a recent experimental study in the aeroacoustic wind tunnel at the Brandenburg University of Technology, both the aerodynamic noise radiation as well as the flow field around a set of porous covered cylinders were investigated. The acoustic measurements were performed at Reynolds numbers (based on cylinder diameter) between 16000 and 100000 using single microphones, while the detailed flow measurements were done using hot-wire anemometry. The results show the influence of the material properties and of the thickness of the porous layer on the aeroacoustic noise and on the turbulence in the wake of the cylinders.
	On the Effects of Leading Edge Serrations on Aeroacoustic Properties during Stall Yannick D. Mayer, Bin Zang, Mahdi Azarpeyvand [Abstract] The understanding of aerofoil noise and possible mitigation approaches is of key importance for the aerospace and wind turbine industry. Despite their relevance and mostly unknown physical noise generation mechanism, separation and stall noise have not received much research attention. In order to control the aerodynamic stall behaviour, serrated leading edges have previously been employed resulting in a less abrupt stall behaviour of aerofoils, but their effects on the hydrodynamic pressure field around aerofoils is not well understood. Hence, the current experimental investigation aims to further the understanding of stall and separation noise and its mitigation using leading edge serrations. Experiments are therefore conducted in the aeroacoustic facility of the University of Bristol using a highly instrumented NACA0012 aerofoil with a large number of static pressure taps and unsteady surface pressure transducers. A variety of leading edge serrations have been designed in order to vary the strength of the induced vortices by the leading edge serrations at high angles of attack. Additionally, simultaneous hot-wire measurements will allow correlation measurements to be conducted between unsteady surface pressure fluctuations and the existing flow field, in order to improve the understanding of the underlying physics.
	Optimisation of Leading-edge Undulations of a NACA 65(12)-10 Aerofoil for Noise Reduction and Aerodynamic Enhancement Kwing-So Choi, Jung-Hoon Kim, Giovanni Lacagnina, Phillip Joseph, Seyed Mohammad Hasheminejad, Tze Pei Chong, Muhammad Farrukh Shahab, Mohammad Omidyeganeh, Alfredo Pinelli [Abstract] Leading-edge undulations or tubercles of humpback whale flippers have been known as one of biomimetic technologies adaptable to flow control of aerofoils, particularly at post stall conditions. These leading-edge undulations are also known to reduce noise resulting from an interaction with on-coming turbulence. We have recently carried out a parametric studiy of a NACA 65(12)-10 aerofoil with a view to optimise the amplitude and the wavelength of leading-edge undulations for noise reduction and aerodynamic enhancement. A 3x3 test matrix composing of three amplitudes (h = 3%, 6% and 12% chord) and three wavelengths (Lambda = 10%, 20% and 30% chord) was used in the investigation, where lift, drag and noise were measured at the Reynolds number of 100,000. In this test, a turbulence-generating grid was installed at the inlet of the test section to increase the turbulence level in the freestream and to promote transition to turbulence near the leading edge of aerofoils without a need for a trip device. Within the test matrix considered, we found that the best improvement in CL,max is given with the greatest wavelength and amplitude, whereas the maximum noise reduction is obtained with h/C = 0.09 and Lambda/C = 0.2.
	Leading- and trailing-edge noise reduction using serrations of new geometry Benshuai Lyu, Lorna Ayton, Chaitanya Paruchuri [Abstract] The leading- and trailing-edge noise due to the interaction of unsteady flow with the sharp edges of aerofoils are common in many engineering applications. One of the most widely used reduction methods is to use serrations, such as those of a sawtooth and a sinusoidal profile. The optimal serration geometry, however, remains largely unknown. In this study, we propose a new type of serration which has an ogee shape. Both model predictions and experimental tests show such a new type serration provides improved performance over conventional sawtooth serrations for leading- and trailing-edge noise reduction. More importantly, this new geometry does not introduce noise increase across the entire frequency range of interest. It is widely known that the leading-edge noise reduction appears to agree more favourably with model predictions than the trailing-edge noise reduction. In this study we also investigate the possible causes for such a difference by closely examining the wall pressure wavenumber spectrum under the turbulent boundary layer.
	Aeroacoustic and Aerodynamic Study of Trailing-edge Serrated Airfoils in Tandem Configuration Xiao Liu, Syamir Alihan Showkat Ali, Mahdi Azarpeyvand, Yannick D. Mayer [Abstract] This paper presents a comprehensive experimental study on the application of trailing-edge serrations as a passive control method for reducing the aerodynamic loading on airfoils in several tandem configurations. The purpose of this study is to investigate the effectiveness of serrated trailing-edge on NACA65-710 tandem airfoil to control and regularize the turbulent flow within the gap area between the two airfoils. The wake flow characteristics for an isolated NACA65-710 airfoil with and without the trailing-edge serration treatment have been carried out using two-dimensional Particle Image Velocimetry(PIV) method. The aeroacoustic performance of the tandem airfoil configuration is also studied using in-site pressure transducers instrumented over the airfoil surface, as well as far- field microphones. To reveal the effects of trailing-edge serration on the tandem configuration, flow experiments were performed using a slotted-sawtooth serration, for several tandem configurations, with different airfoil gap distances. The results show that the use of serrations can generally lead to a significant reduction in the turbulent kinetic energy within the gap, due to the interaction between the flows issued from the tip and root of the serration. The far-field noise results have shown that a significant noise reduction can be achieved with the use of the serrations.
	Numerical and experimental investigation on effects of blade tip-rake on vortex structure and aerodynamic noise of axial-flow fans in an outdoor unit of air-conditioners Seo Yoon Ryu, Cheolung Cheong, Jong-Uk Kim, Byeong Il Park, Se Min Park [Abstract] In the case of axial fans, there are more complex mechanism involved in transferring the energy from the fan blades to a flowing fluid and generating aerodynamic noise due to the blade tip vortex. In this paper, the effects of blade tip shape on vortex structure and aerodynamic noise of axial fans which are used as a cooling fan in an outdoor unit of air-conditioners are numerically and experimentally investigated. Two types of axial fans are considered: one is with blade tip-rake and the other is without blade tip-rake. The blade tip-rake in introduced in the axial fan to reduce the tip vortex as the wing fence does in the airplane. First, the flow field around the fans was predicted by solving the three-dimensional unsteady compressible RANS equations using computational fluid dynamics techniques. Then, the validity of the numerical results are confirmed by comparing the predicted flow rates with the measure ones. The characteristics of flow field driven by the fans with and without blade tip rake are analyzed in detail, especially with an emphasis on the coherent vortex structure. Finally, the aerodynamic noise of axial fans are characterized according to the vortex structure related to blade tip shape.
	Near-field Aeroacoustic Characteristics of a Stalled NACA 0012 Aerofoil Yannick D. Mayer, Bin Zang, Mahdi Azarpeyvand [Abstract] Stall and separation noise can be major contributors to the noise emissions of rotating blades, such as wind turbines, propellers and fans. However, the underlying physics of the noise generation mechanism of stall noise has not received the same research attention as trailing edge noise. Hence, this paper investigates the effects a separated boundary layer has on near-field flow structures and turbulent surface pressure fluctuation statistics, and therefore noise radiation. In order to investigate the wall bounded flow with flow separation, experiments are conducted in the aeroacoustic facility of the University of Bristol using a highly instrumented NACA 0012 aerofoil with 300mm chord. The aerofoil is equipped with 87 static pressure tappings and 91 miniature microphones measuring surface pressure fluctuations over the aerofoil. The experiments are conducted for free-stream velocities between 10m/s and 30m/s and a wide range of angles of attack, including the post-stall regime. In addition, single-wire and cross- wire measurements are conducted synchronously to the unsteady pressure measurements, permitting correlation statics between the collected flow quantities to be determined. This aims to improve the understanding between eddies present in the flow field and the resulting wall pressure statistics, and thereby the noise generation mechanism.
	An Experimental study on Aeroacoustics and Shock Dynamics Associated with Hartmann Whistle Sonu K Thomas, K Srinivasan [Abstract] Hartmann whistle refers to a device that employs a high speed jet impinging on a cavity, which is open at one end and closed at the other. One advantage of using this device is that, it employs no moving part, which in turn makes it more reliable and safe. This device is capable of producing high amplitude tones over a wide range of frequencies. The aero-acoustic field produced due to resonance can have high amplitudes which can be used in Active Flow Control (AFC) applications. Experiments were done to study the interaction between jet and open- closed end cavity, which in turn is responsible for the high amplitude sound produced. The nature of these sound field produced are not completely evident from the prior studies. In the full paper, the aero-acoustic measurements, the resonance induced pressure oscillation measurements in the cavity and the flow visualization using a high speed Schlieren technique will be presented at different nozzle pressure ratios. The main intend of the present work is to understand the underlying physics of such flows interacting with open-closed end cavity, more clearly. This combined measurements can shed more light on the high intensity sound fields thus produced.

Organization:

M. Taroudakis, FORTH-IACM & University of Crete S. Dosso, University of Victoria

	Extraction of Interface Wave Dispersion Curves from Ocean Ambient Noise Hefeng Dong, Guoli Wu, Ganpan Ke [Abstract] Interface wave dispersion measurements have been widely used in geoacoustic inversion to estimate seabed shear-wave speed profile over the past years. This paper extracts interface wave dispersion curves from the cross-correlations retrieved from ocean ambient noise. This technique requires an isotropic distribution of the noise sources in order to ensure reliable travel- time estimates obtained from the cross- correlations. However, real ambient noise fields are often influenced by highly coherent, directional sources generated by ships, earthquakes and other human activities, resulting in a bias of the travel-time estimates. In order to reduce the influence of energetic, directional sources, an eigenvalue-based approach is proposed and applied to the eigen-spectrum of the sample covariance matrix. This approach separates the eigen-spectrum into different components based on the statistical model of the covariance matrix and attenuates directional noise and self- noise related components by a sequential hypothesis tests. This procedure is applied to a dataset of ocean ambient noise recorded by a cable. The total recording time is 1.02 hours with a sampling interval of 2ms. The method shows a clear improvement compared with classical temporal and spectral normalization approaches and makes it possible to obtain reliable dispersion measurements over a short recording period.
	Effect of a soft sediment layer on acoustic normal mode propagation and geoacoustic inversion in the New England Mud Patch Gopu R Potty, James H Miller [Abstract] Previous studies have shown that the travel times of acoustic normal modes corresponding to the Airy Phase regions were found to be extremely sensitive to shear. Simple inversion schemes were developed to estimate the shear speed in the ocean bottom by comparing theoretical predictions with experimental data. Modal dispersion characteristics of broadband data collected during experiments conducted in Middle Atlantic Bight and New England Mud patch were analyzed and bottom shear speeds were estimated. The estimated shear speeds were also compared with shear speeds calculated from core data. In this study the bottom model will be revised to include a soft sediment layer over the elastic basement. The modal dispersion will be calculated using this bottom model corresponding to the New England Mud patch environment using propagation model ORCA. The effect of the addition of soft mud layer on dispersion of lower order modes will be investigated and presented. The sensitivity of modal propagation to the soft mud sediment layer will be explored. [Work supported by the Office of Naval Research].
	Underwater Acoustic Localization of Pulsed Sources with an Array of Three Hydrophones Emmanuel Skarsoulis, Despoina Pavlidi [Abstract] In connection with the development of an acoustic observatory for sperm whales off south Crete an array of three moored hydrophones is considered. It is known that range, depth and bearing of a pulsed source can be estimated using a pair of hydrophones by exploiting the time differences between direct and surface-reflected arrivals. However, that approach suffers from left-right ambiguity, as well as from increased range/depth and azimuth uncertainty for source locations close to the broad side and the endfire of the two hydrophones, respectively. The addition of a third hydrophone, not aligned with the other two, removes the left-right ambiguity and can also lead to reduced uncertainties in range, depth and bearing estimation by increasing the spatial diversity, associated with the different orientation of each hydrophone pair. Furthermore, this configuration allows for range approximation by combining simple bearing estimates based on time differences of direct arrivals at the different hydrophone pairs, e.g. in cases where surface-reflected arrivals cannot be resolved. The presentation will focus on the benefits from the use of the three-hydrophone configuration compared to localization with a pair of hydrophones. [Work partially supported by the Stavros Niarchos Foundation / ARCHERS project].
	Bayesian Geoacoustic inversion for SBCEX17 Reflection, Dispersion, and Ship-noise Data Stan Dosso, Charles Holland, Julien Bonnel, Dag Tollefsen, Josee Belcourt, Jan Dettmer, David Knobles [Abstract] This paper considers a Bayesian inference approach to model selection, parameter estimation, and uncertainty quantification in seabed geoacoustic inversion, with application to data from the 2017 Seabed Characterization Experiment (SBCEX17). The approach is based on trans-dimensional (trans-D) sampling which considers the number of seabed layers as an unknown parameter in the inversion. Trans-D sampling is carried out with the reversible jump Markov chain Monte Carlo algorithm, which constructs a Markov chain that samples from the posterior probability density (PPD) of the geoacoustic model parameters. Wide but efficient sampling of the trans-D parameter space is achieved using principal-component reparameterization and parallel tempering. The data-error model is based on the assumption of multivariate Gaussian errors with correlations represented by an autoregressive process-trans-D sampling of zeroth- and first-order autoregressive processes is applied to avoid over- or under-parameterizing the error model. The Bayesian inference approach is applied to three acoustic data sets frrom the New England Mud Patch during SBCEX17, including wide-angle (spherical-wave) reflection coefficients collected using a moored receiver and towed source; dispersion of water-borne acoustic modes using a single source and receiver, resolved by warping analysis; and matched-field inversion of ship-of-opportunity noise recorded at a bottom- mounted horizontal receiver array.

Organization:

M. Taroudakis, FORTH-IACM & University of Crete S. Dosso, University of Victoria

	Acoustic Signal Characterization using Hidden Markov Models with applications in Acoustical Oceanography. Costas Smaragdakis, Michael Taroudakis [Abstract] The work presents a new probabilistic characterization scheme for acoustic signals with applications in Acoustical Oceanography. This scheme assign the stationary wavelet packet coefficients of the signal, to a representative left-to-right Hidden Markov Model (HMM). In particular, the coefficients of the signal are modeled via a set of Gaussian emission distributions and a matrix of transiting probabilities. The training of the HMM is performed using the Expectation-Maximization algorithm. Similarities measurements between signals then are calculated in terms of the Kullback-Leibler divergence (KLD) by comparing their corresponding HMMs. The applications in Acoustical Oceanography consist of both simulated and experimental data, in all test cases, a single acoustic signal due to a known acoustic source, recorded at a single hydrophone is considered for the retrieval of environmental parameters (ocean acoustic tomography or sea-bed classification applications). The inversion process follows an optimization approach on the basis of KLD, through a Genetic Algorithm. Finally, the results are compared with those obtained using the Statistical Signal Characterization Scheme (SCSS) proposed by Taroudakis et al. [JASA, Vol 119, 1396-1405 (2006)] Moreover, the results corresponding to the real experiment are compared to reference values from the literature.
	Target Detection Method for Reverberant Environments in Continuous-wave Active Sonar System based on Group Multichannel Nonnegative Matrix Factorization Seokjin Lee [Abstract] In the underwater environment, the active sonar system emits an acoustic wave and receives echo signals to detect a target. However, the echo signals consist of not only the echo from the target but also the reverberation from scatterers. If the target has a low speed, the detection problem is more difficult because the target echo is interfered by the reverberation severely. In this paper, the target detection method for the reverberant environment is developed based on multichannel nonnegative matrix factorization. To extract the target echo signal mixed with reverberations, received signal is transformed into bearing-time-frequency domain, and analyzed into bearing, frequency, and temporal basis by the multichannel nonnegative matrix factorization. The simulation was performed to evaluate the proposed method, and the result shows that the proposed method has enhanced detection performance than the conventional methods in the simulated reverberant environments.
	Estimation of bottom parameters using measurement and analysis of sound intensity fluctuations in the presence of moving nonlinear internal waves in shallow water. Boris Katsnelson, Valery Grigorev, Yanyu Jiang, Yun Ren [Abstract] Methodology of estimation of bottom parameters is proposed using data processing and analysis of sound intensity fluctuations due to mode coupling in the presence of moving nonlinear internal waves (NIW) in shallow water channel. As an examples experiments Shallow Water 2006 (Atlantic shelf of USA) and ASIAEX 2001 (South China sea) are considered. In these experiments NIW generated in area of the shelf break were moving approximately along acoustic track with the length about a few tens of km. Spectra of intensity fluctuations of the sound intensity of pulses during a few hours on vertical line array (VLA) due to mode coupling in both experiments have specific peaks corresponding to the difference between wave numbers of coupling pairs of modes. Analysis of fluctuation spectra includes both temporal dependence of fluctuations on separate hydrophones and temporal dependence of intensity summarized over VLA. It is shown that geoacoustic inversion in given case can be achieved using comparison of experimental spectral and temporal dependencies with results of theoretical modeling, containing bottom characteristics (sound speed, density and attenuation) as a fitting parameters.
	Method to determine far-field beampattern of long array from subarray beampattern measurements Donghwan Jung, Jeasoo Kim, Hongju Gu, Sungyong Moon [Abstract] Beampattern measurement is essential to verify the performance of array sonar. However, the classic problem encountered in beampattern measurement of arrays is that the far-field conditions for reaching the plane wave are limited due to reasons such as the measurement space. We aim to measure the beampattern of the array in a limited space and to evaluate the performance through the -3dB beam width. Therefore, we propose a method to measure the beampattern of discrete line array in limited space. The proposed method is based on the sub-array method. A discrete line array whose measurement space does not satisfy the far-field condition is divided into several sub-arrays satisfying the far-field condition, and the beampattern of the line array is derived through the beampattern for the sub-array. The proposed method was verified by simulation and we performed a measurement experiment on a line array of 256 elements with a design frequency of 455 kHz. This method can measure the beampattern of the line array simply as compared with the method of converting the beampattern obtained from the measurement data in the near-field to the far-field.

Organization:

J. Abshagen, WTD 71 C. Zerbs, Müller-BBM GmbH

	Towards a Realistic Approach of Sound Propagation in EIAs: How to Represent Moving Sources and Moving Receivers Uwe Stöber, Lars O. Mortensen, Frank Thomsen [Abstract] Recent guidelines for marine mammal protection like those in the US or in Denmark impose dose-based criteria that require to accumulate the noise received by a marine mammal over time. Compared to a static receiver, fleeing animals receive a significantly lower noise dose because they move away from the zone of noise impacts. Animal movement models can therefore be useful to predict more realistic noise impacts on marine mammals. In addition, noise sources are not always static. During seismic exploration for example, air guns are towed behind a vessel and the locations of the airgun sounds are changing with the forward movement of the vessel. Here we present a feasible solution for noise impact assessments that compromises between realistic representations of mammal behavior and sound field on the one hand and computational resources on the other hand to provide the best possible accuracy in a realistic time frame. The approach consists of two steps: First, the sound field is derived from a number of detailed propagation simulations. Then, an Agent Based Model (ABM) simulates animal movement and noise dose accumulation. Our study shows that noise impact assessments benefit from using a more dynamic approach.
	A Computational Investigation into the Influence of the Shear Properties of the Seabed on Sound Propagation in Shallow Water. Ray Kirby, Wenbo Duan [Abstract] The propagation of sound waves in the ocean is influenced by the acoustic characteristics of the seabed, especially in shallow water. If the seabed is considered to be an elastic or viscoelastic structure, then sound propagation will be influenced by the shear and longitudinal properties of this structure. Accordingly, this paper examines the sensitivity of sound propagation to the shear properties of the ocean floor for shallow waters and when the sound source lies in the ocean. This is motivated by predictions reported in the structural mechanics literature, which demonstrate that wave propagation in viscoelastic structures can be highly sensitive to the shear properties of the material. This investigation uses the semi analytic finite element method to undertake a set of numerical experiments. This numerical approach provides a computationally efficient way of solving the fully coupled acoustic problem, with Bergmann’s equation used for the fluid, and Navier’s elastodynamic wave equation for the seabed. Both elastic and viscoelastic structures are examined and predictions are reported for a range of shear properties, and the relative sensitivity of sound propagation on these shear properties is then investigated.
	Three-dimensional acoustic parabolic equation model based on GPU processing Keunhwa Lee, Woojae Seong, Youngnam Na [Abstract] We introduce a three-dimensional acoustic parabolic equation (PE) model based on GPU processing. This model is developed by splitting the full exponential operator in the three-dimensional split-step marching solution in the Cartesian coordinate. Each exponential operator is approximated by the Taylor series and the sum representation of Pade approximation, which gives an opportunity for GPU parallel processing of the 3D PE solver. An advantage of this model is to implement higher-order cross-terms appearing to the operator approximation efficiently. In the talk, we show the computational performance of GPU based PE code with the comparison of a traditional code for several ocean environments.
	The Analysis of Parabolic Equation Model Solutions using Split-Step/ Finite Difference Method Mustafa Aslan [Abstract] Parabolic equation models solved using the split-step Fourier (SSF) algorithm are commonly studied for the modeling of acoustic wave propagation in the underwater environment. The smoothing function, in the solution of realistic problems, is introduced in SSF algorithm to treat the density change between the water column and the sediment. The algorithm with the smoothing approach accumulates the phase error with range. As an alternative method, by Yevick and Thomson, the hybrid split- step/finite difference method was developed to numerically accomplish the density discontinuity issue and significantly improved the phase error. In the method, the additional operator containing density dependent terms are introduced to implement the hybrid approach. In this work, the results of SSF parabolic equation model with the smoothing and the hybrid approach are compared and are discussed the effect of numerical parameters on the model implementation.
	Parallel implementation for Three-Dimensional Acoustic Field Computation in a Penetrable Wedge by Image Source Method Wenbin Xiao, Yongxian Wang, Wei Liu, Qiang Lan, Xinghua Cheng, Zijie Zhu, Xin Wang, Ben Luo, Dezhi Wang, Jiani Wu, Lilun Zhang [Abstract] Present work aims to the open multi-processing (OpenMP) parallel implementation for the three-dimensional acoustic field computation in a penetrable wedge. Possible practical application of the wedge waveguide propagation is extended to the continental shoulder or shallow water environment. By the source image method, the acoustical propagation solution is in conjunction with an extension of reflection analysis of spherical wave filed from a planar interface. Multiple reflections at the boundaries are taken as the originating from a source image depended on the surface and bottom positions. Analytical expression of acoustical pressure here is coupled with the Bessel function expansion in complex form. Two range-dependent cases, such as the along the slope (two dimension) and cross the slope (three dimension), are tested by the OpenMP parallelization and performance optimization. The multi-core platform of ”Knights Landing” or KNL and many-core platform of ”TianHe-II” are applied to the utilization. The performance gain observed in using OpenMP of the Fortran code on high performance computing facilities is demonstrated.
	Sound propagation experiments in a Norwegian fjord Jan Abshagen, Volkmar Nejedl [Abstract] Vertical stratification of temperature and salinity along with the increase of hydrostatic pressure with depth result in the formation of sound channels in the ocean as well as in deep fjords. A sound channel may form in the surface layer but also in a deep water layer with its acoustical axis being located below the main thermocline. Due to fresh water inflow, mixing, or tidal currents, sound propagation conditions may vary significantly with time but also display substantial variations between different locations, in particular in a fjord. We have performed experiments on sound propagation with RV Elisabeth Mann Borgese in the Norwegian Sognefjord. The measurements were performed with a freely drifting projector buoy and a vertical receiver array, as well as a towed horizontal receiver array in a deep and in a surface channel, respectively. In order to determine the directional behavior of the sound field, beamforming techniques were applied. This work addresses frequency-dependent propagation losses, as well as the spatial, temporal, and directional variability of sound channel propagation in confined waters.
	An Underwater Vehicle Shape with Reduced Acoustic Backscatter Tom Avsic [Abstract] In some underwater applications, acoustic backscatter might have undesired consequences, such as interferences in underwater communications or echoes in the sonar image. Acoustic backscatter of an underwater object is classically reduced by applying appropriate absorbing materials. Absorbing materials convert the acoustic energy into heat and their thickness is usually of the order of a quarter wavelength. In some cases, the application of absorbing materials might have unacceptable drawbacks on other design criteria. For long-distance acoustic propagation (acoustic channel propagation), it can be shown that undisturbed sound travels nearly horizontally within a limited vertical angle of not more than ± 20∘ from the horizontal plane. Hence, for long-distance applications, the acoustic backscatter can be reduced by proper shaping of the vehicle’s lateral outer hull. In case the outer hull is not opaque enough to shield the vehicle’s inner structures (with potentially high backscatter), reflective material can be applied. The working mechanism of reflecting materials is the impedance mismatch and hence, reflecting materials can be much thinner than absorbing materials. This conference contribution will address the worldwide sound spreading in the ocean and show the guidelines for geometric outer hull shaping.
	Development of Underwater Acoustic Transducers Yang Liu, Houlin Fang, Liangyong Zhang, Fang Zhang, Deyu Sun [Abstract] Underwater acoustic transducer which undertakes the task of signal generation and reception is the core equipment for sonar detection. For understanding the mechanism of underwater acoustic transducers, we have reviewed recent typical transducers based on different working principles and different vibration modes, including flextensional transducers, cylindrical radiating transducers, longitudinal transducers and so on. Our research is of great importance for obtaining underwater acoustic transducer with higher performance.
	Analysis of Displacement Amplification Characteristics of Class IV Flextensional Shell Based on Elliptic Perimeter Approximation Formula Houlin Fang, Difeng Sun, Fang Zhang, Yang Liu, Tianqing Zhao, Cheng Zhang, Liangyong Zhang, Xubin Liang, Deyu Sun [Abstract] As the most widely used underwater acoustic transducer, class IV flextensional transducer can get larger amplitude flexural vibration from longitudinal vibration of driven stack by the displacement amplification effect, thereby radiating more energy. In order to further quantitatively understand the displacement amplification characteristics of elliptical tube shell, this paper proposed to calculate the displacement magnification using the ellipse perimeter approximation formula, and the variation law of the displacement amplification with the compression coefficient and the deformation rate was obtained, which was analysed and compared with the results by the method of finite element analysis. The empirical formula for calculating the finite element results by the results of the perimeter approximation formula was got by fitting curves. When the compression coefficient is 0.24∼0.80, the error of the calculation results was not more than 6.11%, and finally the preliminary verification was carried out by experiments. The research of this paper has certain reference significance for the estimation of the radiation performance of low frequency transducers.
	Passive Ship Localization in a Shallow Water Using Pre-trained Deep Learning Networks Dezhi Wang, Lilun Zhang, Changchun Bao, Shuqing Ma, Yongxian Wang [Abstract] Subject to the lack of detailed environmental information, the classical matched-field processing (MFP) may not be adapted to the accurate localization of ocean acoustic sources. In this paper, a framework that applies deep learning techniques instead of the conventional MFP method is presented for the localization of ship acoustic sources in a shallow water environment. The original data in terms of sound pressure is recorded from a vertical array placed in the costal waters. The acquired data is converted into normalized sample covariance matrices (SCMs), which are used as input data fed into the deep-learning architecture. In particular, a framework is proposed to predict the range information of a target ship in a transfer learning manner based on the pre-trained Xception model that is the state-of-the-art convolutional neural networks. The proposed method achieves a performance up to 10 km at range prediction, which is significantly better than that of conventional MFP method and some other shallow machine learning methods. Different from the MFP approach, machine-learning methods are driven directly by the data, which offers an opportunity to enable them to overcome the environmental mismatch problem.

Organization:

A. Homm, Wehrtechnische Dienststelle für Schiffe und Marinewaffen (WTD71) V.F. Humphrey, ISVR, University of Southampton

	Innovative solutions to reduce the transfer of structure borne noise in couplings Lothar Kurtze [Abstract] Modern couplings for ships and especially for mega yachts have to fulfill various tasks. Located between gearbox and propeller shaft, they have to minimize the transfer of structure borne noise while transmitting high torque levels. In addition, they have to resist high misalignments due to very softly mounted frames of ship engine and gearbox. Couplings also have to deal with sometimes difficult torsional vibration situations of the drivetrains, they have to be lightweight and in case of electric motors they often need electric insulation properties. To fulfill all these partly contradictory demands, innovative approaches are essential. In this paper, the development of a lightweight coupling made of a combination of glass and carbon fibers, elastomer layers and steel will be presented. After the design phase, the couplings have been tested at an acoustic test bed. In the next step they have been tested together with an Advanced Electric Drive (4 MW-range). And finally, the new Geislinger Silenco®-coupling has been validated during sea trials. Thus, the coupling helps significantly to reduce the noise radiation. This includes the noise reduction in the ship cabins as well as the noise reduction into the environment.
	Application of the Energy based Finite-Element-Method to determine the sound emission of vibrating ship structures in the High Frequency Domain Henning Lohmann, Boris Dilba, Olgierd Zaleski, Otto Von Estorff [Abstract] For large scale structures the applicability of classical numerical methods such as Finite-Element-Method (FEM) or Boundary- Element-Method (BEM) is restricted to the low and mid frequency range. The fine mesh resolution for higher frequencies rapidly leads to problem sizes which exceed given computational resources. Here, the energy based Finite- Element-Method (EFEM) provides an efficient solution. The EFEM is based on balancing input, output and dissipated powers over a specific control element and is applicable to fluid and structure problems. Exchange of energy between elements is described by transmission coefficients. The unknowns of the EFEM are energy densities which aren’t oscillatory in nature. Thus, the required discretization is independent of the frequency and results in a smaller system of equations compared to classical methods. In this contribution the application of the EFEM to a ship is outlined. After a brief introduction to theoretical aspects the focus will be on the coupling of fluid and structure problems based on the radiation efficiency. In addition to determining the radiation efficiency the influence of the coupling parameters on the resulting system of equations is shown and utilized to define a lower frequency bound for the application of the EFEM to vibrating ship structures.
	Analysis Methods and Design Measures for the Reduction of Noise and Vibration Induced by Marine Propellers Julian Kimmerl, Paul Mertes, Vladimir Krasilnikov, Kourosh Koushan, Luca Savio, Mario Felli, Moustafa Abdel-Maksoud, Ulf Göttsche, Nils Reichstein [Abstract] The paper presents the ongoing work in the research project ProNoVi. The objective of the project is to improve the numerical and experimental methods for the prediction of noise and vibrations induced by a propeller operating behind ship hull in full-scale conditions. Based on the improved methods practical recommendations for the reduction of noise and vibration levels for different classes of vessels are a further objective. The propeller cavitation noise is identified as the dominating noise source, which coincides with important frequencies of perception of marine fauna and thus may have negative impact on marine life. As a side effect, reducing the noise under water will increase comfort and thus safety for crew and passengers on board. The project aims at delivering a better understanding of fundamental physical mechanisms related to turbulence, induced vorticity and cavitation dynamics, which play a decisive role in generation of tonal and broadband propeller noise. The project partners will give a joint presentation, focusing on state-of-the-art in the field, today’s challenges in both the experimental and numerical approaches, and how the ProNoVi project is addressing those. ProNoVi is funded by the European Union. Further information is available at www.martera.eu/projects/pronovi.
	Numerical modelling of the correction factor for predicting the monopole source level from shallow water noise measurements on vessels Victor F Humphrey, Yin Cen [Abstract] Environmental issues have led to a growing interest in the measurement of the radiated underwater noise from surface vessels. ANSI and ISO have supported this by issuing standards for the measurement of the Radiated Noise Level of vessels in deep water. In addition, a draft standard for the conversion of such measurements into the Monopole Source Level, required as an input to environmental noise models, is at an advanced stage. However, many operators and investigators are constrained to making measurements in shallow water where both sea-surface and seabed effects are important. A numerical study has been performed in order to understand the variability of the correction factor required to convert measurements of received pressure into the equivalent Monopole Source Level in such an environment. These calculations have been performed using an image source model and have included the effects of reflection at both the sea-surface and an attenuating seabed. The influences of the water depth, seabed type, distance of closest approach, and number and location of the hydrophones have been investigated. The results show how the correction factor varies and indicate the extent to which simplified approximations can be used for the correction factor.
	Measurements of Underwater Noise from Pile Driving in Southwest Coast of Korea Dong-Gyun Han, Daehyeok Lee, Jee Woong Choi [Abstract] The interest in underwater noise produced by impact pile driving has been increasing worldwide. Since the impact pile driving noise with extremely high sound pressure level can cause negative effects on marine ecosystem, it is important to measure the noise and to assess possible risks. Recently, offshore wind power generation is being actively developed in Korea and accordingly efforts to assess the environmental impact of pile driving noise are being made. Measurements of pile driving noise were carried out twice in the southwest coast of Korea in 2017 and 2018. The variation of the impact pile driving noise with the number of strikes was analyzed for the noise measured at a single site. In addition, impact pile driving noise were measured as a function of source-receiver range, which was compared to the theoretical transmission loss curves predicted by a damped cylindrical spreading model and the parabolic equation-based range dependent acoustic model (RAM).
	Shipping noise propagation at the shallow sea Eugeniusz Kozaczka, Grażyna Grelowska [Abstract] The issue of rapid noise growth in the oceans and seas is currently a challenge for research teams dealing with underwater pollution. At present, for the EU water areas, a directive has been developed, which is intended to cause a reduction of acoustic background, the main component of which is underwater noise mainly related to maritime transport. The paper proposes a method to determine the noise map on the shallow sea basin based on the classical solution of the wave equation for specific boundary conditions. As a result, we get a solution consisting of a superposition of analytical solutions for the issue of own waves related to the depth in relation to changes in the distance to source. For a stationary problem, the expression describing the noise of the selected vessel in spectral form will be used as a function of excitation. This formulated issue allows us to estimate the underwater noise level at any distance and depth from the source. The analysis of the superposition of underwater acoustic disturbances makes it possible to determine the value of the sound pressure level from the sources of these disturbances. Details of the full procedure will be presented in the presentation.
	Noise mitigation for the construction of offshore wind turbines – Overview of Noise Mitigation Systems for complying with noise limits. Michael A. Bellmann, Robert Kühler, Rainer Matuschek, Patrick Remmers, Siegfried Gündert, Michael Müller, Jan Schuckenbrock [Abstract] Underwater noise caused by impact pile-driving during the installation of offshore foundations is potentially harmful to marine life. Therefore, in several nations (e. g. Germany, The Netherlands, Denmark, Taiwan, etc.), noise limiting values were developed to prevent injury of marine life. The experience over the last years shows, that underwater noise produced during pile-driving, depending on many parameters and measurements, and shows values, which exceed the national underwater noise limiting value by up to 20 dB. Therefore, Noise Mitigation Systems (NMS) are required to significantly minimize the underwater noise. Therefore, many different noise mitigation systems were applied during the last years during installation of Offshore foundations and were accompanied by underwater noise measurements in accordance to the ISO 18406. In this paper, a general overview of the existing and tested NMS, including various tested system configurations as well as combinations of different NMS, will be presented. Additionally, the achieved overall resultant noise reductions during pile driving as well as the factors influencing the noise reduction will be discussed. Moreover, it will be explored what implications the development towards larger foundations can have for the state of the art in noise mitigation.
	Uncertain Parameters of the Propagation Path in 3D Pile Driving Noise Modelling Jonas Von Pein, Johannes Seidel, Elin Klages, Stephan Lippert, Otto Von Estorff [Abstract] Nowadays, offshore constructions are mainly mounted to the sea bed by piles. These piles are driven into the sea floor by percussive pile driving. This process leads to high underwater sound pressure levels. Therefore, limiting values have been set in many countries for the protection of the marine fauna. In order to get a profound prognosis of the emitted sound levels, numerical models are essential. These models are mainly rotationally symmetric and based on a finite-element formulation for the close vicinity of the pile and a far-field propagation model for ranges further away than 1 km. In the current contribution the far-field model is a full 3D parabolic equation model. A major problem in pile driving noise radiation modelling is to derive a reliable description of the acoustical properties of the sea bottom. However, due to the inhomogeneous character of the soil these parameters are always subjected to variability. To quantify the effects of this uncertainty in the acoustical soil parameters, a sensitivity analysis is performed for a pile driving scenario assuming a heavily changing bathymetry with and without noise mitigation. Furthermore, the sensitivity analysis is performed with Nx2D and 3D simulations.
	Evaluation of range standards for underwater radiated noise of ship’s in beam aspect. Hans Hasenpflug, Anton Homm, Layton Gilroy, Stefan Schäl [Abstract] The underwater radiated noise levels of a small naval vessel were measured at the Heggernes deep water sound range. The data evaluation and analysis was performed in accordance with the NATO standard STANAG 1136 and the ISO 17208-1 standard. Both standards describe procedures for the determination of underwater radiated noise levels of ships in beam aspect on a deep water sound range. To assess and compare the results of both standards, the hydrophone configuration at the Heggernes range was utilized at various depth settings. Measurements were made of the ship in two different ship conditions (machinery states) and a variety of sailing speeds. Some of these measurements were repeated a large number of times. The resulting 3rd octave based spectra are used to compare the two procedures.
	Structure-borne Noise Reduction of Gearboxes in Maritime Application Robin Daniel Seiler [Abstract] The reduction of vibration and noise levels on board ships offers not only occupational safety for the crew, but also increased comfort for passengers and less impact on the environment. In order to meet predefined noise and vibration requirements, an effective noise control concept must integrate a variety of aspects and includes several coordinated counter measures. The foundation for this is laid in the engine room, because the excitement of the ship’s structure in the engine room decisively shapes the vibro-acoustic situation on board. Here you can ”gain” the most decibels. Single or double-elastic suspensions for ship engines already show high transmission losses, when rubber mounts are used. As a result, the structure-borne noise via gearboxes becomes a crucial sound transmission path. Against this background, the central role of gearboxes in powertrains of ships is investigated. The presented countermeasures include different drive train configurations: 1. Highly flexible couplings with low torsional stiffness between engine and gearbox, 2. Propeller shaft couplings in combination with very soft gearbox mounting systems, 3. Active vibration controls to reduce velocity amplitudes on the gearbox foundations. The effects of the outlined solutions on the reduction of airborne noise levels in receiving rooms on board are discussed.