Residual Data Vector Method of Underwater Acoustic Source Localization by a Three-Dimensional Array

Abstract

Three-dimensional (3-D) localization of multiple acoustic sources in the ocean is a problem of abiding interest. Existing methods of 3-D localization such as MUSIC and Capon processor involve a computationally complex 3-D search for locating the peaks of an ambiguity function of three variables. Normally, a two-dimensional (2-D) array of sensors, composed of a horizontal linear array (HLA) and a vertical linear array (VLA) with a sufficiently large vertical span, is deployed for making the acoustic field measurements. In this paper, we propose a computationally simpler method that requires J searches each in 1-D space and 2-D space for 3-D localization of J sources. Simulation results, using measurements by a 3-D X-shaped array (3DXA), are presented to illustrate the performance of the proposed method. It is shown that measurements by a 3DXA with a short vertical span can provide a satisfactory range-depth estimation performance that is normally attained using measurements by a VLA with a much larger span.