Volume 115, Issue 2, February 2004
Index of content:
- UNDERWATER SOUND 
115(2004); http://dx.doi.org/10.1121/1.1635417View Description Hide Description
The van Cittert–Zernike theorem is used to generate models for the spatialcoherence of a sound field that has been forward scattered from the sea surface. The theorem relates the spatialcoherence of an observed wave field to the distant source intensity distribution associated with this field. In this case, the sea surface upon ensonification is taken to be the source, and the sea-surface bistatic cross section corrected for transmission loss is taken as a surrogate for the source intensity distribution. Improvements in methodology for generating an estimate of the 2D autocorrelation function for sea surface waveheight variation, necessary to compute the bistatic cross section, are documented in the Appendix. Upon invoking certain approximations, simple expressions for the characteristic length scales of vertical, horizontal, and horizontal–longitudinal coherence, are derived from the theorem. The three coherence length scales identify a coherence volume for the spatialcoherence of a sound field arriving via the surface bounce channel. Models for spatialcoherence derived from the van Cittert–Zernike theorem without these approximations compare reasonably well with measurements of complex vertical coherence made at 8 kHz and 20 kHz in the East China Sea as part of the 2001 ASIAEX field program. In terms of the ASIAEX field geometries and sea-surface conditions, at frequency of 20 kHz the coherence volume is a vertical layer 0.5 m thick by 3 m in each of the two horizontal dimensions; at 8 kHz these dimensions increase by a factor of 2.5, representing the ratio of the two frequencies.
115(2004); http://dx.doi.org/10.1121/1.1640389View Description Hide Description
Recently, adaptivity was introduced to time-reversal mirror to steer the nulls, and referred to as an adaptive time-reversal mirror (ATRM) [J. S. Kim, H. C. Song, and W. A. Kuperman, J. Acoust. Soc. Am. 109, 1817–1825 (2001)]. In this study, ATRM is extended to simultaneous multiple focusing in an oceanwaveguide. The multiple focusing is achieved by imposing a set of constraints in the formulation to find the weight vectors. The algorithm is applied to the long-range underwater acoustic communication to show, via simulation, that the simultaneous pulse compression at multiple receiving locations is achieved.
115(2004); http://dx.doi.org/10.1121/1.1636760View Description Hide Description
The use of adjoint modeling for acoustic inversion is investigated. An adjoint model is derived from a linearized forward propagation model to propagate data-model misfit at the observation points back through the medium to the medium perturbations not being accounted for in the model. This adjoint model can be used to aid in inverting for these unaccounted medium perturbations. Adjoint methods are being applied to a variety of inversion problems, but have not drawn much attention from the underwater acoustic community. This paper presents an application of adjoint methods to acoustic inversion. Inversions are demonstrated in simulation for both range-independent and range-dependent sound speed profiles using the adjoint of a parabolic equationmodel. Sensitivity and error analyses are discussed showing how the adjoint model enables calculations to be performed in the space of observations, rather than the often much larger space of model parameters. Using an adjoint model enables directions of steepest descent in the model parameters (what we invert for) to be calculated using far fewer modeling runs than if a forward model only were used.
115(2004); http://dx.doi.org/10.1121/1.1634277View Description Hide Description
This work concerns the problem of estimating the depth of a submerged scatterer in a shallow-water ocean by using an active sonar and a horizontal receiver array. As in passive matched-field processing (MFP) techniques, numerical modeling of multipath propagation is used to facilitate localization. However, unlike passive MFP methods where estimation of source range is critically dependent on relative modal phase modeling, in active sonar source range is approximately known from travel-time measurements. Thus the proposed matched-field depth estimation (MFDE) method does not require knowledge of the complex relative multipath amplitudes which also depend on the unknown scatterer characteristics. Depth localization is achieved by modeling depth-dependent relative delays and elevation angle spreads between multipaths. A maximum likelihood depth estimate is derived under the assumption that returns from a sequence of pings are uncorrelated and the scatterer is at constant depth. The Cramér–Rao lower bound on depth estimation mean-square-error is computed and compared with Monte Carlo simulation results for a typical range-dependent, shallow-water Mediterranean environment. Depth estimation performance to within 10% of the water column depth is predicted at signal-to-noise ratios of greater than 10 dB. Real data results are reported for depth estimation of an echo repeater to within 10-m accuracy in this same shallow water environment.