Using deconvolution to suppress range sidelobes for MIMO sonar imaging

Abstract

A multiple- input multiple-output (MIMO) sonar can produce a large-aperture virtual array with the use of a small size array and a limited number of elements, which is very useful when applied to underwater acoustic imaging. However, the range sidelobes (SLs) in the MIMO sonar imaging result is high due to the non-ideal orthogonality of transmitting waveforms. In this paper, we propose a deconvolution method to suppress the range SLs. We consider that the range SLs in the imaging result are composed of cross-correlation functions (CCFs) and SLs of auto-correlation functions (ACFs) of transmitting waveforms. And they can be treated as the “noise” term in the output of the convolution between the original distributions of scatterers and a Dirac-like point spread function (PSF). Accordingly, we use the deconvolution method (i.e., the Richardson-Lucy algorithm) to suppress these range SLs. We give the basic processing procedure of the proposed deconvolution method. Simultaneously, we give a simple searching method to find the best PSF corresponding to the lowest range SL level after deconvolution. Numerical simulations and a tank experiment show the proposed deconvolution method can suppress the range SLs effectively.