Using Low-Complexity Adaptive Beamformer to Improve MIMO Sonar Imaging Performance

We present an improved MIMO sonar imaging method utilizing the low-complexity adaptive (LCA) beamformer. LCA beamformer can be regarded as a simplified version of the minimum variance distortionless response (MVDR) beamformer or an adaptive extension of the conventional beamformer (CBF). Although both MVDR and LCA aim to minimize the noise and interference in the imaging process, MVDR calculates the optimal beamformer weight from the spatial statistics of the wave field, while LCA selects the best weight from a set of pre-defined weights. In this paper, we use a set of Kaiser and Chebyshev windows for LCA to choose from. The Kaiser windows are used to suppress sidelobes (SLs), while the Chebyshev windows are used to improve the resolution. Numerical simulations demonstrate that the proposed MIMO sonar imaging method using LCA beamformer outperforms the one using CBF in terms of resolution and SLs. Compared with MVDR, LCA is more robust and has lower computational complexity, albeit with weaker performance than MVDR.