Sparse Bayesian Learning for Horizontal Wavenumber Retrieval in Underwater Acoustical Signal Processing

Abstract

According to the classical normal mode theory, low frequency acoustical signals propagating in shallow water are composed of several modal components associated with their own horizontal wavenumbers. In this paper, the horizontal wavenumbers are retrieved by the sparse Bayesian learning approach using a vertical line array. The modal depth functions derived from the Beam-Displacement-Ray-Mode theory are used as the dictionary. The proposed method does not require the prior of sea bottom information (e.g., soud speed). The performance is demonstrated by simulations in a shallow water environment.