Active Sonar Target Imaging and Classification System

Active sonar-based target classification of submarines, minisubmarines, mines, and swimmers is very important in littoral warfare. Current active sonar systems are inadequate because they require many emissions per potential target, yield high false alarm rates, and suffer from high clutter interference. New models of active biosonar signal analysis based on bat-like target range profiling and dolphin-like image construction have shown the potential to eliminate or significantly reduce these problems. ORINCON has joined with Raytheon to form a uniquely qualified team to develop this biosonar signal processor and VLSI implementation. The team combines extensive signal processing, active sonar, biosonar, neural network, and VLSI experience. The specific aims of our proposed Phase I research are: (1) implementation of a number of biologically motivated algorithms, (2) preparation of a number of data sets for use in algorithm comparisons, (3) evaluation of the performance of the various algorithms, (4) selection of the best algorithm, and (5) performance of a feasibility analysis for a VLSI implementation. Phase II efforts will focus on the design, implementation, test, and evaluation of a VLSI chip of the selected biosonar algorithm.