Synthetic adaptive matched field processing for moving source range estimation in deep water.

Abstract

Adaptive matched field processing (AMFP) has proven effective for source localization in deep-water environments. However, when the target is in motion, the need for numerous snapshot samples can lead to distortion in covariance estimation, degrading AMFP performance. A synthetic AMFP method has been proposed to compensate for the phase of multi-snapshot signals, enhancing AMFP performance. Additionally, a rough estimation of target velocity is obtained. The efficacy of the method has been validated through numerical simulations and experimental data, with results showing that, within a 9 km range, the average localization error is reduced by 1.45 km compared to traditional AMFP.