Index of content:
Volume 117, Issue 2, February 2005
- ACOUSTIC SIGNAL PROCESSING 
Intersymbol interference in underwater acoustic communications using time-reversal signal processing117(2005); http://dx.doi.org/10.1121/1.1841692View Description Hide Description
Coherent underwater communication is hampered by the time spread inherent to acoustic propagation in the ocean. Because time-reversal signal processing produces pulse compression, communications has been suggested as a natural application of the technique. Passive versions of time-reversal processing use a receive-only array to do combined temporal and spatial matched filtering. It can be shown, however, that the pulse compression it achieves is not perfect and that an equalizer that relies solely on time-reversal processing will have an error floor caused by uncompensated intersymbol interference (ISI). In the present paper, a physics-based model is developed for the uncompensated ISI in a passive time-reversal equalizer. The model makes use of a normal-mode expansion for the acoustic field. The matched-filtering integral is approximated and the intermediate result interpreted using the waveguide invariant. After combining across the array and sampling, formal statistical averages of the soft demodulation output are calculated. The results show how performance scales with bandwidth, with the number and position of array elements, and with the length of the finite impulse response matched filters. Good agreement is obtained between the predicted scaling and that observed in field experiments.
Characterization of subwavelength elastic cylinders with the decomposition of the time-reversal operator: Theory and experiment117(2005); http://dx.doi.org/10.1121/1.1811471View Description Hide Description
The decomposition of the time-reversal operator provides information on the scattering medium. It has been shown [Chambers and Gautesen, J. Acoust. Soc. Am. 109, 2616–2624 (2001)] that a small spherical scatterer is in general associated with four eigenvalues and eigenvectors of the time-reversal operator. In this paper, the 2D problem of scattering by an elastic cylinder, imbedded in water, measured by a linear array of transducers is considered. In this case, the array response matrix has three nonzero singular values. Experimental results are obtained with linear arrays of transducers and for wires of different diameters smaller that the wavelength. It is shown how the singular value distribution and the singular vectors depend on the elastic velocities the density ρ of each wire, and on the density and velocity of the surrounding fluid. These results offer a new perspective towards solution of the inverse problem by determining more than scattering contrast using conventional array processing like that used in medical ultrasonic imaging.
117(2005); http://dx.doi.org/10.1121/1.1847894View Description Hide Description
Multipath arrivals at a receiving sensor are frequently encountered in many signal-processing areas, including sonar,radar, and communication problems. In underwater acoustics, numerous approaches to source localization, geoacoustic inversion, and tomography rely on accurate multipath arrival extraction. A novel method for estimation of time delays and amplitudes of arrivals with maximum a posteriori (MAP) estimation is presented here. MAP estimation is optimal if appropriate statistical models are selected for the data; implementation, requiring maximization of a multidimensional function, is computationally demanding. Gibbs sampling is proposed as an efficient means for estimating necessary posterior probability distributions, bypassing analytical calculations. The Gibbs sampler includes as unknowns time delays, amplitudes, noise variance, and number of arrivals. Through Monte Carlo simulations, the method is shown to have a performance very close to that of analytical MAP estimation. The method is also shown to be superior to expectation-maximization, which is often applied to time-delay estimation. The Gibbs sampling approach is demonstrated to be more informative than other time-delay estimation methods, providing complete posterior distributions compared to just point estimates; the distributions capture the uncertainty in the problem, presenting likely values of the unknowns that are different from simple point estimates.
117(2005); http://dx.doi.org/10.1121/1.1850405View Description Hide Description
An enhanced simulated annealing algorithm is used to invert sparsely sampled seismic data collected with sonobuoys to obtain seafloor geoacoustic properties at two littoral marine environments as well as for a synthetic data set. Inversion of field data from a 750-m water-depth site using a water-gun sound source found a good solution which included a pronounced subbottom reflector after 6483 iterations over seven variables. Field data from a 250-m water-depth site using an air-gun source required 35 421 iterations for a good inversion solution because 30 variables had to be solved for, including the shot-to-receiver offsets. The sonobuoy derived compressional wave velocity–depth models compare favorably with models derived from nearby, high-quality, multichannel seismic data. There are, however, substantial differences between seafloor reflection coefficients calculated from field models and seafloor reflection coefficients based on commonly used regression curves (gradients). Reflection loss is higher at one field site and lower at the other than predicted from commonly used gradients for terrigenous sediments. In addition, there are strong effects on reflection loss due to the subseafloor interfaces that are also not predicted by gradients.