Application of smoothing techniques to sequential geoacoustic inversion

Abstract

Sequential Bayesian techniques such as particle filters have been successfully used to track a moving source in an unknown, complex, and evolving ocean environment. These methods treat both the source and the ocean parameters as non-stationary unknown random variables and sequentially estimate the best solution in addition to the uncertainties in the estimates. Particle filters are numerical methods called sequential Monte Carlo techniques that can operate on nonlinear systems with non-Gaussian probability density functions. Particle smoothers are a natural extension to the filters. A smoother is appropriate in applications where all data have already been observed and are readily available. Therefore, both past and future measurements can be exploited. Geoacoustic and source tracking is performed here using two smoother algorithms, the forward backward smoother and the two-filter smoother. The approach is demonstrated on experimental data collected during both the SWellEx-96 and SW06 experiments where the uncertainty in the estimates is reduced.