Acquisition concepts for MIMO sonar

The objective of sonar processing is to enhance the signal received from the target with respect to noise or clutter in order to enhance detection, classification and localisation. Range-bearing information is commonly obtained through beamforming and matched-filtering of data acquired with a densely sampled detector array. The source sampling, on the other hand, is usually sparse, since data processing is typically performed using single ping data. In this paper we discuss techniques that can be used to increase the source sampling in sonar applications. This is of interest since beamforming with both sources and detectors potentially enhances the resolution of the beamforming and the signal-to-noise ratio in comparison to beamforming with detectors only. This can be achieved either with multiple sources, or by digitally combining data acquired at different times. Examples are continuous transmission, blended acquisition, source coding for multiple sources in a multistatic setting, and the virtual source technique. We illustrate these techniques with seismic data examples and discuss the prospects for their use in sonar applications