Adaptive filter of seabed clutter onboard the AUVs of an active multistatic sonar network

Abstract

This work aims at reducing the false alarm rate of a multistatic active sonar working in coastal, shallow waters. The sonar system is implemented through a heterogeneous manned-unmanned network including autonomous underwater vehicles (AUVs) equipped with receive hydrophone arrays. According to a cognitive sonar philosophy, each AUV characterizes the environment by processing in real time the acquired acoustic data, and exploits this contextual information to optimize the sonar performance in terms of increased probability of detection and reduced false alarm rate. The paper presents an algorithm to identify areas of compact clutter on the seabed in order to filter out as false alarms the contacts falling in those areas. Promising experimental results are reported, which were obtained under different environmental conditions (during LCAS16 and DMON17 sea trials).