Multichannel blind deconvolution of sound source of opportunity in ocean waveguide

Abstract

Signals that travel through an ocean waveguide are typically distorted when they are received by a remote receiver because of interference and distortion arising from multiple propagation paths. This presentation investigates the applicability of a physics-based blind-deconvolution technique proposed by Sabra et al. (JASA 127, EL42, 2010) to sound sources of opportunity such as randomly radiating ships. Using only the noisy recorded signals from an underwater receiver array, this blind deconvolution provides a means to estimate the Green's functions between the source of opportunity and the elements of the receiver array as well as the original signal radiated by the source of opportunity. These estimated Green's functions can then be used for acoustic characterization of the ocean waveguide parameters which is typically done using controlled sources only. We will discuss the performance of the proposed approach using both numerical simulations and at-sea data using discrete shipping events recorded on a vertical array as sources of opportunities.