Coherent space‐time‐frequency processing to enhance the bistatic detection and classification of elastic shells in shallow water.

Abstract

For underwater sonar, time‐frequency analysis has been shown to be a powerful tool for detection and classification of man‐made targets. For instance, with traditional monostatic systems, a key energetic feature of spherical shell is the coincidence pattern, or midfrequency enhancement, that results from antisymmetric Lamb‐waves propagating around the circumference of the shell. The development by the Navy of mine countermeasure sonar systems, using a network of autonomous systems, provides a mean for multiple bistatic measurements, and thus potentially bistatic enhancement for target detection. However, time‐frequency representations of bistatic simulations of scattered signals from spherical shell show that this coincidence pattern typically shifts in both time and frequency with respect to the monostatic case. Hence, this time‐frequency shift is challenging for bistatic target detection algorithms based on standard array processing techniques. Design of robust multistatic sonar system based on the generalized space‐time‐frequency coherence of the bistatic measurements will be discussed. The influence of the source‐receiver configuration and interface reflections on the proposed approach have been investigated numerically and experimentally using data collected in shallow water with an elastic spherical shell [Work supported by ONR Code321, N000140810087.]