A Hybrid Algorithm for Sparse Antenna Array Optimization of MIMO Radar

Abstract

A sparse array is designed to achieve a narrower beam without increasing hardware costs. To reduce the peak side-lobe level and the grating lobe in the desired ambiguity-free region, a hybrid algorithm combining particle swarm optimization (PSO) and convex optimization is presented. While the main beam scans in the desired region, the positions and excitation of array elements are alternately optimized by PSO and convex optimization, respectively. The simulation results show that, compared with the PSO algorithm alone, the hybrid algorithm obtains the lower peak side-lobe level. Furthermore, the side lobes remain almost the same in the desired region when the beam scans. A multiple-input multiple-output (MIMO) radar prototype equipped with the designed sparse array is presented. The experimental result shows that two closely placed moving targets can be separated in two-dimension (2-D) by the MIMO system, which certificates the effectiveness of the proposed algorithm in actual scenarios.