A real-time cognitive-sonar system for diver detection

Cognitive signal processing has been intensively studied over the past years. Nevertheless, cognitive sonar systems only draw little attention in the research community. In this paper the signal processing of a cognitive sonar system for diver detection is shown in detail. It is inspired by the target-detection system of a bat and thus able to adapt its signal processing units to measured parameters of particular targets. This is done by adjusting the transmitted signal as well as the direction of the transmitted energy of the active sonar system by transmit beamforming. Thereby, it is possible to improve on the one hand the probability of detection of a target for a given signal-to-noise ratio (SNR) as well as signal-to-reverberation ratio (SRR) and, on the other hand, to extend the spatial awareness by adapting the transmit beamformer. This technique leads to a combination of two spatial filters: the receive beampattern and the transmit beampattern. To obtain target-specific parameters, the data form an internal tracking system is used. After showing a general system overview, the core parts of the signal processing are shown in detail. At this point we focus on the Doppler processing with its real-time interpolation and correlation algorithms. MonteCarlo simulations illustrate the potential of the proposed system by comparing the results with conventional active-sonar signal processing.