Volume 106, Issue 5, November 1999
- acoustics research letters online
- acoustical news—usa
- book reviews
- general linear acoustics 
- nonlinear acoustics 
- aeroacoustics, atmospheric sound 
- underwater sound 
- ultrasonics, quantum acoustics, and physical effects of sound 
- structural acoustics and vibration 
- noise: its effects and control 
- architectural acoustics 
- acoustic signal processing 
- physiological acoustics 
- psychological acoustics 
- speech production 
- speech perception 
- speech processing and communication systems 
- music and musical instruments 
- bioacoustics 
- letters to the editor
Index of content:
- ACOUSTICS RESEARCH LETTERS ONLINE
106(1999); http://dx.doi.org/10.1121/1.427181View Description Hide Description
Knowing the response of miniature ultrasonichydrophones at frequencies below 1 MHz is important for assessing the accuracy of acoustic pressure pulse measurements in medical ultrasound applications. Therefore, a time delay spectrometry (TDS) system was developed as an efficient means to measurehydrophone sensitivity in this frequency range. In TDS a swept-frequency signal is transmitted. A tracking receiver distinguishes arrivals with different propagation delays by their frequency offset relative to the signal being transmitted, thus eliminating spurious signals such as those reflected from the water surface or tank walls. Two piezoelectricceramic source transducers were used: a standard planar disk and a disk with varying thickness to broaden the thickness-resonance. This latter design was preferred for its more uniform response without significant sensitivity loss. TDS is not an absolute method, but it was demonstrated to provide efficient, accurate calibrations via comparison with a reference hydrophone using a substitution technique.
Intelligibility of 1/3-octave speech: Greater contribution of frequencies outside than inside the nominal passband106(1999); http://dx.doi.org/10.1121/1.427606View Description Hide Description
We reported previously that “everyday” sentences were highly intelligible when limited to a 1/3-octave passband centered at 1,500 Hz and having transition-band slopes of approximately 100 dB/octave. The present study determined the relative contributions to intelligibility made by the passband (PB) and the transition bands (TBs) by partitioning the same bandpass sentences using 2,000-order FIR filtering. Intelligibility scores were: PB with both TBs, 92%; deletion of both TBs (leaving only the 1/3-octave PB with nearly vertical slopes), 24%; deletion of the PB (leaving both TBs separated by a 1/3-octave gap), 83%. These and other results indicate a remarkable ability to compensate for severe spectral tilt and the consequent importance of considering frequencies outside the nominal passband in interpreting studies using filtered speech.
- GENERAL LINEAR ACOUSTICS 
106(1999); http://dx.doi.org/10.1121/1.428071View Description Hide Description
A new impulse-response model for the edge diffraction from finite rigid or soft wedges is presented which is based on the exact Biot–Tolstoy solution. The new model is an extension of the work by Medwin et al. [H. Medwin et al., J. Acoust. Soc. Am. 72, 1005–1013 (1982)], in that the concept of secondary edge sources is used. It is shown that analytical directivity functions for such edge sources can be derived and that they give the correct solution for the infinite wedge. These functions support the assumption for the first-order diffraction model suggested by Medwin et al. that the contributions to the impulse response from the two sides around the apex point are exactly identical. The analytical functions also indicate that Medwin’s second-order diffraction model contains approximations which, however, might be of minor importance for most geometries. Access to analytical directivity functions makes it possible to derive explicit expressions for the first- and even second-order diffraction for certain geometries. An example of this is axisymmetric scattering from a thin circular rigid or soft disc, for which the new model gives first-order diffraction results within 0.20 dB of published reference frequency-domain results, and the second-order diffraction results also agree well with the reference results. Scattering from a rectangular plate is studied as well, and comparisons with published numerical results show that the new model gives accurate results. It is shown that the directivity functions can lead to efficient and accurate numerical implementations for first- and second-order diffraction.
106(1999); http://dx.doi.org/10.1121/1.428072View Description Hide Description
The existence of the surface wave in an upward refracting medium has been predicted by Raspet et al. [J. Acoust. Soc. Am. 89, 107–114 (1991)]. Making use of an acoustic analogy, one is able to study the surface waves above a convex cylinder and, hence, to simulate the propagation of sound over an impedance ground in an upward refracting medium. A suitable comblike surface may be used to facilitate the generation of surface waves above the convex ground. This paper describes laboratory experiments for the observation of surface waves over a convex impedance ground.
106(1999); http://dx.doi.org/10.1121/1.428073View Description Hide Description
Diffraction of sound by concave surfaces is investigated theoretically and experimentally. In an earlier study [J. Acoust. Soc. Am. 104, 2683–2691 (1998)], it has been demonstrated that a rigorous analogy exists for the sound field above a convex circular cylinder in an otherwise homogeneous medium. The predicted sound field corresponds to the situation where the sound speed of the medium decreases exponentially with height. Extending the previous work, this paper investigates of the sound field above a concave surface and explores the corresponding analogy. Normal mode solutions have been developed for a downward refracting medium with an exponential sound speed profile. The solutions are used to predict the sound fields diffracted by a cylindrical concave surface. A series of laboratory experiments is conducted using point monopole, horizontal dipole, and vertical dipole sources over cylindrical concave surfaces. The experimental measurements are compared with the normal mode predictions. For monopole and horizontal dipole sources, good agreement has been found between measurements and the normal mode predictions using an exponential profile. However, the agreement is less satisfactory where the sound field was due to vertical dipole sources.
106(1999); http://dx.doi.org/10.1121/1.428074View Description Hide Description
Measurements of infrasonicseismo-acoustic ambient noise using an ocean bottom seismometer in shallow water have uncovered an unusual phenomenon: the noise spectrum of the horizontal component of seabed velocity shows several prominent peaks in the frequency range 0–8 Hz, whereas the noise spectra of both the acoustic pressure and the vertical component of seabed velocity show very weak or nonexistent features at the same frequencies. This structure is interpreted theoretically as resonances of shear waves of vertical polarization in the upper sediment layer, excited by the diffuse infrasonicsound field in the water. Independent interface wave dispersion studies at the site have revealed an approximate power-law profile of shear speed versus depth, having the form with and (SI units). The theoretical development concentrates on exact analytic solutions for the resonance frequencies and wave field for power-law profiles and on the WKB and more advanced asymptotic solutions in the more general case of smooth shear-speed profiles with a power-law singularity. The experimental observations are interpreted in light of these analytic results, and are consistent with the previously determined power-law shear speed-profile.
106(1999); http://dx.doi.org/10.1121/1.428075View Description Hide Description
This paper presents the boundary conditions that apply to the weak integral formulation of the Biot mixed poroelasticity equations. These boundary conditions are derived from the classical boundary conditions of the Biot displacement poroelasticity equations. They are applied to the surface integrals of the associated weak form to account for exterior excitations, supports, and couplings with exterior elastic, acoustic, poroelastic media, and a septum. It will be shown that the derived boundary conditions for the formulation lead to simpler finite elementequations compared to those obtained from the formulation. Finally, two numerical examples are presented to validate the poroelastic-septum coupling condition, and to highlight the limitations of the free edge condition on a poroelastic medium.
Exact solution for sound propagation in ducts with an axial mean temperature gradient and particulate damping106(1999); http://dx.doi.org/10.1121/1.428127View Description Hide Description
An exact solution for one-dimensional sound propagation in ducts in the presence of axial mean temperature gradient and particulate damping is presented in this paper. The acoustic waveequation is derived starting from the one-dimensional momentum and energy equation. The application of appropriate transformations leads to an analytically solvable Whittaker’s differential equation for the case of a linear mean temperature gradient and Bessel’s differential equation for the case of an exponential mean temperature gradient. The derived analytical solutions are used to investigate the dependence of the acoustic field in a duct on temperature gradient and particulate damping.
106(1999); http://dx.doi.org/10.1121/1.428076View Description Hide Description
Transient scattering from closed rigid bodies can be analyzed using a variety of time domain integral equations, e.g., the Kirchhoff integral equation and its normal derivative. Unfortunately, when the spectrum of the incident field includes one or more of the resonance frequencies of the corresponding interior problem, the solutions to these time domain integral equations become corrupted with spurious interior modes. In this article, this phenomenon is demonstrated via numerical experiments, and a Burton–Miller-type time domain combined field integral equation is proposed as a remedy. To verify that the solutions to this Burton–Miller-type equation are not corrupted by interior modes, various numerical results are presented. It is anticipated that this equation, when used in conjunction with fast time domain integral equation solvers (e.g., plane wave time domain algorithms), will enable the accurate analysis of transient wavescattering from acoustically large bodies.
Fast transient analysis of acoustic wave scattering from rigid bodies using a two-level plane wave time domain algorithm106(1999); http://dx.doi.org/10.1121/1.428077View Description Hide Description
It is well known that the computational cost associated with the application of classical time domain integral equation methods to the analysis of scattering from acoustical targets scales unfavorably with problem size. Indeed, performing a three-dimensional scatteringanalysis using these methods requires operations, where denotes the number of basis functions that model the spatial field distribution over the surface of the scatterer and is the number of time steps in the analysis. Recently, novel plane wave time domain algorithms that augment these classical methods and thereby reduce their high computational cost have been introduced. This paper describes such a plane wave time domain algorithm within the context of the analysis of acoustic scattering from rigid bodies and outlines its incorporation into a time domain integral equation solver in a two-level setting. It is shown that the resulting scheme has a computational complexity of Examples comparing the accuracy and computational efficiency of the conventional and accelerated methods are presented. The proposed two-level scheme renders feasible the broadband analysis of scattering from large and complex bodies.
Elucidation of the relationship between complex acoustic power and radiation efficiency for vibrating bodies106(1999); http://dx.doi.org/10.1121/1.428070View Description Hide Description
This investigation examines the physical meaning of surface complex acoustic power and its relationship to acoustic radiation efficiency. It is shown that the radiated power is the power radiating out of a far-field surface where the plane wave relationship between pressure and particle velocity holds. Meanwhile, the reactive power pertains to the difference between kinetic energy and potential energy. A stationary condition of the ratio between the radiated power to the reactive power yields an eigenvalue problem, subsequently decomposing the surface acoustics into a modal representation. Doing so further allows the examination of the relationship between acoustic radiation efficiency and power factor of the complex power. According to the results, the radiation efficiency of the first radiation mode is nearly equal to the square of the first modal power factor. The modes beyond the first of the modal radiation efficiencies are relatively larger than the corresponding squared modal power factors. Numerical examples of elastic structures subjected to external forces illustrate the relationship between radiation efficiencies and power factors of complex powers.
- NONLINEAR ACOUSTICS 
106(1999); http://dx.doi.org/10.1121/1.428078View Description Hide Description
The problem of convergence of finite difference (FD) schemes for simulating the propagation of pulses in nonlinear media is analyzed. In order to establish the reliability of the numerical procedure, different FD schemes are considered and compared among themselves, with the results of a Von Neumann analysis and with the analytical solution (when available). Examples of application of the method are also provided: e.g., the propagation of semi-infinite waves and pulses in active media with periodic and exponentially decaying nonlinearities.
106(1999); http://dx.doi.org/10.1121/1.428079View Description Hide Description
The Doppler effect is a familiar phenomenon in linear acoustics. The nonlinear Doppler effect is also possible when a moving target exhibits nonlinear responses. In these cases Doppler frequency shifts arise at harmonics or combination frequencies in the scattered waves. For collinear geometry of primary beams, the nonlinear Doppler frequency shifts at the second harmonic and the combination frequencies are found to be the same as those for the linear scattering of those frequency components by the target. However, for highly, noncollinear primary beams, the difference frequency Doppler frequency shift can greatly exceed the linear one. This phenomenon is mainly studied in the present work. Experimental verification done with flowing bubbles in a tube shows good agreement with the theory. The suggested difference frequency Doppler technique can find applications in medicaldiagnostics of blood flow as well as in industries and in oceanology.
- AEROACOUSTICS, ATMOSPHERIC SOUND 
106(1999); http://dx.doi.org/10.1121/1.428128View Description Hide Description
It has been found that the presence of a bias flow can enhance the low-frequency sound absorption of a perforated liner. Therefore it is important to understand the acoustic characteristics of a liner with bias flow for practical applications. This paper presents experimental investigations of the acoustic properties of a perforated liner with bias flow. First, the experimental results show that the presence of a bias flow markedly increases both the absorption coefficient and effective bandwidth of a perforated liner. Second, it is shown in the present experiment that plate thickness has a major influence on the acoustic properties of a liner with bias flow. A simple model is presented to consider this behavior; the model gives good agreement with the measured data. Considering that blowing and suction will have a different effect on the mean shear flow over the surface of a perforated liner, comparative experimental results between blowing and suction are also included in the present investigation.
- UNDERWATER SOUND 
106(1999); http://dx.doi.org/10.1121/1.428080View Description Hide Description
The robustness of a ray travel-time approach is studied using simulated data in both shallow waters and Arctic oceans.Effects of travel-time measurement error, sound speed profile mismatch, and source/array parameters mismatch on the solution accuracy are quantified. The approach is then employed to analyze the data generated by the first two SUS shots in the ICESHELF-89 experiment. Source location, array depth, array tilting angle, sound speed profile, and bathymetry are characterized.
Application of waveguide invariants to analysis of spectrograms from shallow water environments that vary in range and azimuth106(1999); http://dx.doi.org/10.1121/1.428124View Description Hide Description
The waveguide invariant summarizes in a single scalar parameter the dispersive propagation characteristics for a given environment. In this paper, an expression is derived for the invariant in shallow water environments where the bottom bathymetry varies with range and azimuth. It then is applied to the interpretation of broadband, single-element spectrograms collected during SWellEx-3, a shallow water acoustics experiment off the coast of San Diego. The unusual behavior of the interference patterns seen in the spectrograms from this experiment during the transmission of a 75–150-Hz pseudo-random noise sequence by a source towed over range-varying bottom bathymetry is well matched by the model predictions.
Analysis of high-frequency acoustic scattering data measured in the shallow waters of the Florida Strait106(1999); http://dx.doi.org/10.1121/1.428081View Description Hide Description
Experiments were conducted in the Florida Strait region of the United States in July 1995 to investigate the contribution of sediment volume scattering to measured acoustic backscatter levels at frequencies of 7.5 and 15 kHz. Five sites were analyzed for bottom backscattering level in the radial and azimuthal directions. Sediment geophysical parameters were established by a cross-well tomographic measurement and used as inputs for modeling. Both sediment volume and rough interfacescattering models were utilized in analyzing the data. Comparison of model predictions with measured data showed that for grazing angles, at the water–sediment interface, from the critical angle to approximately 60 degrees, volume scattering dominated. For angles greater than approximately 60 degrees, surface roughness scattering consistently dominated. In geographic areas where the interface sound speed ratio exceeded one, roughness scattering controled the backscattered level for grazing angles less than the critical angle. Inversion of the measured acoustic backscatter data was performed using a genetic algorithm optimization developed for use with the volume scattering model. Inversion results were found to agree well with measured data and measured environmental parameters that describe the scattering volume.
106(1999); http://dx.doi.org/10.1121/1.428082View Description Hide Description
The application of a prototype 12-kHz sidescan sonar suitable for horizontal imaging of fish distributions is described. The sonar consisted of a 40-element (20λ) array transmitting a sweep with correlation processing of the received echoes. Observations are reported for two different implementations: (i) a configuration in which the sonar was towed at ∼35 m in water of depth 60–220 m in the southern Strait of Georgia, British Columbia, where the primary target was sockeye salmon, and (ii) a bottom-mounted deployment in a shallow (10–14 m) channel near Copenhagen, Denmark, where herring were present. In the Strait of Georgia, discrete fish targets were detected with up to 25-dB signal-to-reverberation ratio at ranges from 1 to 7 km. In the shallow Danish waters, herring schools were detected at ranges up to 1.2 km. In each case an acoustic propagationanalysis identifies the useful detectioncharacteristics. In the Strait of Georgia the signal is confined to a subsurface sound channel, requiring additional knowledge of vertical distributions of salmon in order to make abundance estimates. In contrast, fish detectability is nearly depth independent in the shallow Danish waters. In the Danish deployment the sonar could be swept in azimuth, providing a sequence of sector images that reveals the changing shape and movement of herring schools. These preliminary results provide an indication of the potential value and limitations of intermediate range sidescan surveys for fish stock assessment.