Adaptive array processing for wide-band active sonars

A high-frequency sector scan sonar can be used to detect and classify man-made objects located on the sea floor. If the hydrophone array of the sonar is to be mounted on a miniaturized autonomous underwater vehicle, the aperture size of the array must be small enough so that it will not affect the vehicle's maneuverability. However, a small aperture size of the array results in poor angular resolution when using the conventional delay and sum beamforming technique. This paper shows that, with the aid of spatial resampling, narrow-band adaptive beamforming algorithms can be applied to improve the angular resolution of wide-band active sonars with a small aperture uniform linear hydrophone array. Owing to the motion of the vehicle and the highly nonstationary environment, only single-snapshot adaptive beamforming algorithms can be used. In this paper, spatial smoothing is used in conjunction with the minimum variance distortionless response and multiple signal classification adaptive beamforming algorithms to produce a high-resolution image using a single ping. Real sonar images generated using these two algorithms are compared with those generated using another single-snapshot adaptive beamforming algorithm, spatial processing: optimized and constrained, and the conventional delay and sum beamforming algorithm.