Biologically inspired processing of radar and sonar target echoes

Abstract

Modern radar and sonar systems rely on active sensing to accomplish a variety of tasks, including detection and classification of targets, accurate localization and tracking, autonomous navigation and collision avoidance. Bats have relied on active sensing for over 50 million years and their echolocation system provides remarkable perceptual and navigational performance that are of envy to synthetic systems. The aim of this study is to investigate the mechanisms bats use to process echo acoustic signals and investigate if there are lessons that can be learned and ultimately applied to radar systems. The basic principles of the bat auditory system processing are studied and applied to radio frequencies. A baseband derivative of the Spectrogram Correlation and Transformation (SCAT) model of the bat auditory system, called Baseband SCAT (BSCT), has been developed. The BSCT receiver is designed for processing radio-frequency signals and to allow an analytical treatment of the expected performance. Simulations and experiments have been carried out to confirm that the outputs of interest of both models are “equivalent”. The response of the BSCT to two closely spaced targets is studied and it is shown that the problem of measuring the relative distance between two targets is converted to a problem of measuring the range to a single target. Nearly double improvement in the resolution between two close scatterers is achieved with respect to the matched filter. The robustness of the algorithm has been demonstrated through laboratory measurements using ultrasound and radio frequencies (RF). Pairs of spheres, flat plates and vertical rods were used as targets to represent two main reflectors.