Analysis and comparison of spatial correlation and intensity distribution statistics

Abstract

Active sonar processing systems operating in shallow-water littoral environments are often plagued with excessive false alarms, generally referred to as sonar clutter. At times the effects of these clutter events can be mitigated through repeated measurements and/or careful sonar operator examination; however, this process can be time consuming and even experienced sonar operators may have difficulty identifying targets in cluttered environments. Additionally, increasing interest in autonomous vehicle platforms and the push for real-time decision and performance management introduces further requirements on sonar systems. This work provides an empirical analysis of the potential in using spatial correlation of the backscattered wavefront as a classification metric, and compares this metric to a commonly used probability density function (PDF) processing method. The shape of the PDF of the backscattered acoustic intensity is often used in sonar processing systems such as constant false alarm rate (CFAR) detectors. These PDFs generally provide an estimate of the Rayleigh-like (random phase scattering) or non-Rayleigh-like (generally finite bright scatterers) nature of temporal variations in acoustic returns. This talk will outline information that may be available about clutter events based on measurements of spatial correlation. [Work supported by Office of Naval Research.]