Ultra-high resolution, biologically inspired sonar

Abstract

Present operational capability to detect and classify mines in adverse environments such as shallow water is very limited. Research has shown that mammalian sonars are superior to any man-made sonars in such environments. This paper describes a novel signal processing concept based on recently discovered echo processing operations used by echolocating bats and dolphins. The signal processor uses unique front-end filters followed by nonlinear functions emulating auditory neural models to effect high resolution at low frequency. Processing emulations confirm performance far superior to conventional processing techniques. Of particular importance is the system implementation using a scalable architecture that can be integrated into an AUV to suit various missions. Results of simulation and data analysis are included here which demonstrate the processing gains realizable by the proposed technique.