Volume 89, Issue 5, May 1991

A review of otoacoustic emissions
View Description Hide DescriptionOtoacoustic emissionsmeasured in the external ear canal describe responses that the cochlea generates in the form of acoustic energy. For the convenience of discussing their principal features, emitted responses can be classified into several categories according to the type of stimulation used to evoke them. On this basis, four distinct but interrelated classes can be distinguished including spontaneous, transiently evoked, stimulusfrequency, and distortionproduct otoacoustic emissions. The present review details the findings that have been described for each emission type according to this classification schema. Additionally, the known features of emitted responses are discussed for both normally hearing and hearingimpaired humans and experimental animals, and with respect to their potential clinical applications. The findings reviewed here clearly indicate that future studies of otoacoustic emissions will significantly increase our understanding of the basic mechanisms of cochlear function while, at the same time, provide a new and important clinical tool.

Numerical implementation of an adaptive fastfield program for sound propagation in layered media using the chirp transform
View Description Hide DescriptionUsing the chirpztransform, a numerically efficient adaptive integration algorithm for fastfield programs has been developed. The fastfield program (FFP) is a numerically efficient algorithm for computation of the sound pressure due to a time harmonic point source above a general boundary in a layered medium. The new algorithm allows the user to specify the precise location of the range sample points, or “detectors,” independent of the horizontal wavenumber sampling grid. This feature makes the algorithm particularly attractive for use in applications where sampling on a predetermined spatial is required. An example of such an application is the computation of the acoustic frequency response as a function of range. In this case the complex sound pressure must be computed at many frequencies while maintaining a constant range sampling interval. In addition, the algorithm can be used to recompute the sound pressure field with different range resolutions without recomputing the horizontal wavenumber spectrum samples. It is also shown that the algorithm can be used to adaptively increase the number of integration points required to evaluate the Sommerfeld integral that results from the FFP type of formulation.

Equation of state of the system of magnetoacoustic waves
View Description Hide DescriptionMagnetoacoustic waves are considered as a system. Bose–Einstein statistics are applied to find the equation of state and thermodynamic quantities of the system. It is found that energy and pressure of the system are directly proportional to the cube of absolute temperature. Adiabatic changes of the system are also discussed.

Diffraction of a cylindrical acoustic wave by an absorbing halfplane in a moving fluid
View Description Hide DescriptionAcoustic diffraction from a line impulse by an absorbing half plane in the presence of a moving fluid is considered. The field is calculated using spatial and temporal Fourier transforms. The results for a finite pulse are also presented. The physical interpretation of the results is then discussed.

Coupled perturbation series for disparate mixtures
View Description Hide DescriptionA technique is presented which has been developed for calculating fields in twophase mixtures when the material properties of the two phases are very dissimilar. A pair of coupled integral equations are derived and then solved together using an alternating recursion. This gives rise to a pair of coupled perturbation series (corresponding to the field point being in one or the other of the two phases). Truncations of the series then lead to descriptions of the solution fields which are valid in the regime of very small (but not necessarily zero) material property ratios. The specific problem of acousticwave fields due to a point source in a twofluid mixture is treated in full and serves to illustrate the procedure. The results are applied to the case of planewave interaction with a flat plane interface, where it is shown that the perturbation series do indeed produce expansions of the exact solutions.

Cross sections of generalized Rayleigh scatterers
View Description Hide DescriptionThis paper highlights the radically different wavelength dependences of the ratio of scattering to geometric cross section for stiffness, resistance, and masscontrolled scatterers characterized by their bulk modulus and structural damping. The relation between solutions formulated by various authors, notably between Rayleigh’s and Lamb’s mutually exclusive results, is explored. Finally, the generalized Rayleigh scattering solution is illustrated for a cavity in a viscoelastic matrix.

The acoustic scattering by a submerged, spherical shell. II: The highfrequency region and the thickness quasiresonance
View Description Hide DescriptionThis article presents a fundamentally oriented analysis of the complexka plane pole structure of the S matrix for spherical shells in the thickness range (a is the outer radius of the sphere) for the highfrequency region (100<ka<1000); the Poisson’s ratio for these shells is fixed at σ=0.34 (aluminum). Surprisingly, this analysis reveals that the most dominant feature in the backscattered form function of a thin spherical shell is a single prominent highfrequency peak. The underlying modal structure and accompanying regions of negative group velocity are related to this feature, which is interpreted as a thickness “quasiresonance.” In particular, the significance of stationary values of the group velocity to the relative amplitude of modal excitation is emphasized.

On the Lynch variational method for scattering of waves by randomly rough surfaces
View Description Hide DescriptionA modification of the physical opticssurface source density based on the Lynch variational approach is proposed for the case of scattering by onedimensional pressurerelease Gaussian random rough surfaces. Rather than the local radius of curvature, it involves the surface correlation length and introduces new criteria validating the physical optics evaluation of the coherent and incoherent scattering intensities. Closed expressions for the coherent reflection coefficient and the incoherent scattering cross section are derived without resorting to highfrequency approximation techniques. Numerical data based on these expressions and conventional geometrical optics shadowing theory are presented and shown to be in very good agreement with recently published exact Monte Carlo simulation results.

On the representation of cavity fields in terms of laterally guided waves
View Description Hide DescriptionLateralwave representation of forced acoustic oscillations in a homogeneous gas contained within a perfectly rigid cavity are compared for two cases that differ only by the presence in one case of a rigid inner core in addition to the rigid outer shell of the cavity. The two representations exhibit increasingly similar laterally propagating modes as the core of the one case is made smaller relative to the wavelength of the oscillation, although significant differences persits near the mode cutoffs. When the core is present, the propagating modes are accompanied by a discrete spectrum of evanescent lateral modes. The evanescent modes become increasingly dense as the size of the core is reduced, to be replaced entirely by a continuous evanescent spectrum in the absence of the core. This continuous spectrum, which is shown in the present instance not always to be negligible, arises mathematically in a manner similar to the “background” continuum found in lateralwave decompositions of the scatter of acoustic waves from penetrable cylinders and spheres.

Synthesis of acoustic matching networks by discrete space Fourier transform method
View Description Hide DescriptionThis paper describes a method used to analyze and synthesize acoustsic planewave matching structures between a transducer and a medium of arbitrary characteristic impedance. The analysis is applicable to multilayer structures with periodic discontinuities in characteristic impedance. It is shown that the frequency response of such structures is approximately equal to the discrete Fourier transform of the individual layer reflection coefficients. The synthesis procedure uses the inverse discrete time Fourier transform to generate a sequence of layer reflection coefficients, and applies a boundary condition to calculate the characteristic impedances. Using this method, wide bandwidth matching structures can be obtained using finite impulse response (FIR) filter topologies as a model. A numerical example is presented to show how the procedure is implemented, and a linear analysis is performed which models the layers of the synthesized structure as ideal transmission lines. The results of the analysis are in good agreement with the response predicted by the Fourier method.

Sound field in a rectangular cavity in the presence of a thin, flexible obstacle by the integral equation method
View Description Hide DescriptionThis paper is concerned with the prediction of the sound field in a cavity enclosing a thin, flexible obstacle and having finite impedance boundaries. The formulation uses the integral equation method adapted to consider a thin obstacle. It also makes use of the equation of motion of the obstacle, but the only form given here is for the case of a limp panel. The solution is achieved by expressing all the required functions as modal expansions in the equations defining the problem. The set of eigenfunctions of an empty, rigidwalled rectangular cavity is used as a basis for the expansions, which restricts the final solution to rectangular cavities. A few numerical results are presented to validate the theory on some aspects related to simple cases of plane obstacles in rigidwalled cavities.

Acoustical tweezers
View Description Hide DescriptionA stable force potential well was generated by two collimated focused ultrasonic (3.5 MHz) beams propagating along opposite directions. Latex particles (270μm diameter) and clusters of frog eggs were trapped in the potential well. The trapped object can be moved axially or laterally by moving one of the PZT focusing transducers that generate the ultrasonic focused beams. The axial position of the trapped object can also be maneuvered by tuning the frequency of the electrical voltage applied on the transducers.

Acoustic propagation in the western Greenland Sea frontal zone
View Description Hide DescriptionA study, utilizing numerical experiments, of the environmental acoustic effects of the western Greenland Sea Frontal Zone in the summer of 1983 along a 185km west to east transect is presented. The transect begins at the edge of the marginal ice zone in the East Greenland Current, then crosses the East Greenland Polar Front and terminates in the water of the Greenland Sea gyre. The transect contains several mesoscale eddylike features and a sharp front between Return Atlantic Intermediate Water and Greenland Arctic Intermediate Water, whose effects on an acoustic field have not been previously reported. This front can impart >15dB increase in loss in <10 km. The effects of the frontal zone on acoustic propagation as a function of location and depth of the acoustic source and depth of the receiver for a nominal frequency of 50 Hz are discussed. Significant differences, sometimes greater than 20 dB, in the acoustic field were observed and are explained in the analysis for variations in any of the source and receiver parameters. The soundspeed structure on either side of the frontal zone makes propagation loss and the dominant acoustic modes very dependent on direction and location. The direction of propagation with the lowest loss (as much as 15dB difference along the entire transect) can change with a modest shift in source from a 10m depth to 150 m. As compared to the nonfrontal environment, propagation in the presence of the frontal zone, for any of the source and receiver depths studied, can result in changes in the levels and distribution of energy in the acoustic field by as much as 15 dB as well as changes in the mode of propagation (e.g., refracted/surface reflected to SOFAR). It can also result in an increase (>10 dB) in the sensitivity of propagation loss to changes in receiver depth (10–150 m) compared to the nonfrontal zone environment. A significant finding is that a longer transmission range (93 km) through the waters of a frontal zone can reduce the effects of crossing the front itself by as much as 15 dB. It was also observed that 75 km of upslope propagation that included two oceanic fronts was unaffected by a decrease in bottom depth of 2000 m within the 75km range.

Azimuthal variation of lowfrequency acoustic propagation through asymmetric Gulf Stream eddies
View Description Hide DescriptionA numerical modeling study of the azimuthal variation of 25Hz acoustic propagation through radially asymmetric Gulf Stream eddies is presented. Both a coldcore (cyclonic) and a warmcore (anticyclonic) eddy, which are quasielliptical in shape, are investigated. Sources located both inside and outside the eddies are considered with transmission out to 240 km. The source is at a depth of 150 m where there are substantial environmental variations due to eddies. Receivers are at depths of 100, 150, and 300 m. The dimensions of the first four convergence zones (CZ) are utilized as indicators of the azimuthal variation of propagation. Propagation conditions assumed an absorbing bottom and no azimuthal coupling of energy. Sensitivity to change in azimuth is dependent upon the orientation of the eddy relative to the direction of transmission. It is increased for transmission paths parallel to the major axis relative to those parallel to the minor axis. For propagation parallel to the major axis, a 15° change in azimuth can result in a change of as much as 13 km in width and 15 km in range for the third CZ. An irregularity in the shape of a quasielliptical eddy is shown to effect an appreciable difference in transmission. A 30° shift in the direction of propagation to the right side of the major axis of a coldcore eddy produced a 20% change in the width of the third CZ, while a 30° shift to the left side of the axis gave a 64% change in the width. Sensitivity can depend upon the direction of propagation along a major axis for a source located within the eddy. Propagation in one direction can produce a sensitivity for a 20° change in azimuth of more than double the shift in range to the third CZ, as the opposite (180°) direction yielded over a 75° change in azimuth.

Lowresolution acoustic scattering models: Fluidfilled cylinders and fish with swim bladders
View Description Hide DescriptionLow spatial resolutionacoustic scatteringmodels were developed to model the scattering by objects that are less ideal than perfect cylinders, i.e., fish. The models used the first few modes of solutions for finite and bent fluid cylinders [T. K. Stanton, J. Acoust. Soc. Am. 83, 55–63 (1988) and 86, 691–703 (1989)]. Empirical mode attenuation factors and bent cylinders were used to control the spatial resolution. All computations were for backscatter in the far field. First, numerical evaluations were made for frequency domain scattering amplitudes. The scattering amplitudes were inverse Fourier transformed to time domain for identification of the reflection and diffraction components. Gasfilled cylinders represented fish swim bladders. Fish bodies were represented by a liquidfilled cylinder where the liquid had a density and a sound velocity a little greater than water. Gasfilled cylinder: The sum of the zeroth and first mode gave a reflection from the front face and a diffraction from the boundary of the insonified and shadow zone. Summations over 23 modes gave a single reflection and no boundary diffracted wave for ka<20, where k is wave number and a is cylinder radius. Liquidfilled cylinder: The sum of the zeroth and first mode gave reflections from the front and back faces and shadow boundary wave. The shadow boundary wave was larger than the reflected components. The sum of 23 modes gave reflections from the front and back face and no shadow boundary wave for ka<20. Fish model: The theoretical models were compared to measurements of sound scattered by anchovy ([R. H. Love, J. Acoust. Soc. Am. 49, 816–823 (1971) and D. V. Holliday, J. Acoust. Soc. Am. 51, 1322–1332 (1972)]. The matches of data and theory were excellent. Comparison of the bent cylinder theory and measurements to the traditional spherical bubble model showed that the spherical bubble was a poor approximation to the swimbladders of these fish. The research shows that acoustic modeling needs fish length, fish mass, fish and swimbladder morphology, etc.

GLORIA interference patterns with modes akin to surfaceduct modes
View Description Hide DescriptionObservations with the GLORIA sonar in coastal waters have shown regular patterns in the reverberation. These are interpreted as due to the interference between the first two normal modes, having characteristics similar to those in the surface duct, but with insonification below the thermocline. Up to four peaks have been seen at a spacing of about 3 km, in good agreement with expectation for the 6.4kHz frequency. The main scatterers appear to be shallow biological targets such as fish, probably located close to the thermocline.

A comparison between the boundary element method and the wave superposition approach for the analysis of the scattered fields from rigid bodies and elastic shells
View Description Hide DescriptionThe steady state analysis of the scattering of plane acoustic waves from submerged rigid and elastic bodies using two approaches is presented. The first approach uses a combined finite element/boundary element (FE/BE) methodology. The NASA structural analysis (NASTRAN) program is used to formulate the structural matrices based on the finite element method(FEM). The surface integral equation radiation and scattering (SIERRAS) program creates the fluid matrices based on the boundary element method(BEM) and solves the coupled fluidstructure interaction problem. A superparametric boundary element (BE) with nine nodes is employed. The combined Helmholtz integral equation formulation (CHIEF) is employed to provide a unique solution for all frequencies. In the second approach, the superposition method (SUP) is used for modeling the fluid. The SUP method is an offboundary approach that employs a number of point sources moved inside the body to represent the fluid response at the surface. This allows the fluid matrices to be formed without surface integration. Formulations for the SUP method are introduced for both the radiation and scattering problems. The program SUPER reads the NASTRAN structural matrices and solves the combined FE/SUP fluidstructure equations. The FE/BEM and FE/SUP methods are applied to the scattering of an infinite set of plane waves from a submerged rigid sphere, rigid prolate spheroid, right cylinder, and from an elastic cylindrical shell with hemispherical endcaps. The SUP method is found to be easier to implement and to provide an accurate result at internal resonance frequencies where the BEM requires additional equations to ensure an accurate solution.

Horizontal spatial correlation of bottom reverberation for normal incidence
View Description Hide DescriptionThe spatial correlation in the backscattered field for waves incident normally on a random boundary is derived using Kirchhoff–Helmholtz scattering theory in which the effects of both boundary heights and slopes are included. The acoustically rough case is considered and it is shown that the scattering process may be interpreted in terms of a deep, random phase screen; the influence of the boundary slopes is interpreted as diffraction at the screen. For a directive source, it is shown that the spatial correlation distance of the field is determined by the correlation of the source shading for a wide variety of boundary conditions; limited experimental data supports this result. For an omnidirectional source, the spatial correlation coefficient is

Bistatic reverberation calculations using a threedimensional scattering function
View Description Hide DescriptionThe ocean bottom scattering function depends, in general, on the grazing angles and the azimuthal angles of the incident and scattered energy. However, most measurements are for backscatter only. The few general measurements that are available indicate strong forward scattering near the angle of the specularly reflected ray and weaker, azimuthally isotropic, diffuse scattering away from the specular angle. By combining Lambert’s law scattering with a surface scattering function based on the Kirchhoff approximation, Ellis and Haller [J. Acoust. Soc. Am. Suppl. 1 82, S124 (1987)] proposed a function that incorporated these features. The function is quite simple, and depends on three parameters that can be fitted to backscattermeasurements. The functional form thus allows a reasonable extension from backscatter to the general threedimensional scattering function, which can then be used in bistatic reverberation calculations. It is an improvement over two commonly used methods (which do not include azimuthal dependence) for extrapolating backscattering to general scattering: the separable approximation, and the halfangle approximation. This paper discusses the threedimensional function in more detail, and presents some comparisons between model predictions and measured bistatic reverberation.

Acoustical radiation from thermally stressed sea ice
View Description Hide DescriptionObservations of the soundgenerated by crackingice in the Canadian Arctic during a cooling period provide an opportunity for studying the response of different types of ice to thermal stress. Synthetic aperture radar(SAR) images and air photographs of ice around the experimental site allow mapping of cracking events back to their points of origin, thus providing a correlation between ice type and acoustical activity. It is shown that there are only a few cracking events in the firstyear ice during periods of air temperature variation, whereas multiyear ice produces a large number and is responsible for most of the ambient sound recorded. Analysis of several individual events including both natural and artificial sources reveals that failure processes cause acoustic emissions with an angularly dispersive radiation pattern in which higher frequencies tend to be radiated downward and lower frequencies radiated toward the horizontal. This pattern can be explained in terms of two types of eigenmodes used in plate vibration theory. The theory is extended to the case of a shallow surface crack representative of thermal stress effects. Comparisons are made between theoretical predictions and icecracking events, including artificial sound sources, and the results interpreted in terms of seaice properties.