Extraction of striations from continuous active sonar (CAS) data

Abstract

Previous work has shown that the waveguide invariant can provide a new physics-based tracking constraint that exploits the multipath structure to limit the possible target tracks in pulsed active sonar (PAS) data. The innovation of the invariance extended Kalman filter (IEKF) tracking algorithm is to make predictions and measurements of the target's propagation signature in addition to the conventional quantities of the target displacement and velocity vectors. A necessary preprocessing step is to isolate the target return from the direct arrival. Striation-based beamforming [1] can be used as an optional pre-processing step to preserve the multipath structure in the beamformer output of a horizontal line array (HLA). The recent trend toward continuous active sonar (CAS) processing has generated interest in how the waveguide invariant might be similarly exploited for tracking when the sonar is operated with a high duty cycle. It is expected that the physics-based signal processing algorithms discussed here will extend the applicability of the invariance principle to CAS systems and provide enhanced tracking performance.