Superresolution Multitarget Parameter Estimation in MIMO Radar

This paper is concerned with a multiple-input-multiple-output (MIMO) radar operating in an environment with two or more closely located targets. In this scenario, mutual target interference is a serious problem for multitarget parameter estimation, reducing the performance of existing methods such as least squares, Capon, and amplitude and phase estimation. In contrast to previous methods where the overall effect of mutual target interference is treated as "noise," in this paper, two methods for suppressing this interference are proposed. The first is based on a constrained optimization problem that provides an iterative method. The second involves a novel nonlinear optimization based on a cost function of targets' directions which is solved using the biogeography-based-optimization algorithm. The performances of both the proposed approaches are evaluated via computer simulation studies and shown to outperform existing methods.