Abstract
H hybrid approach based on statistical signal characterization and a linear inversion scheme for the estimation of range dependent sound speed profiles of compact support in shallow water is presented. The approach is appropriate for ocean acoustic tomography when there is a single receiver available, as the first stage of the method is based on the statistical characterization of a single reception using wavelet transform to associate the signal with a set of parameters describing the statistical features of its wavelet sub-band coefficients. A non-linear optimization algorithm is then applied to associate these features with range-dependent sound speed profile in the water column. This inversion method is restricted to cases where the range dependency is of compact support. At the second stage a linear inversion scheme based on modal arrivals identification and a first order perturbation formula to associate sound speed differences with modal travel time perturbations is applied to fine tune the results obtained by the optimization scheme. A second restriction of this stage is that mode identification is necessary. If this assumption is fulfilled the whole scheme may be applied in ocean acoustic tomography for the retrieval of three-dimensional features, combining inversion results at various slices.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call