Bi-iterative MVDR Beamforming based on Beamspace Preprocessing for MIMO radars

Abstract

In this paper, a bi-iterative minimum-variance distortionless response (MVDR) beamforming technique based on the beamspace preprocessing is proposed to solve two typical problems of the traditional MVDR beamformers in multi-input multi-output (MIMO) radars. Firstly, the omnidirectional power is radiated by traditional MIMO radars, however, the interested target is usually located in a certain spatial area. The unnecessary radiation on other areas will reduce the output signal-to-interference-plus-noise ratio (SINR) of the MVDR beamformer. To solve this problem, a transmit beamspace preprocessing by second order cone programming (SOCP) is utilized to focus the energy in the interested region. Secondly, the MVDR beamformer in MIMO radars usually requires a great number of training samples and very high computation load, since both the transmitted degrees of freedom (DoF) and the received DoF are fully utilized. To tackle with these problems, a bi-iterative MVDR beamforming algorithm is proposed based on the beamspace preprocessing. Simulation results demonstrate the efficiency of the proposed beamforming algorithm.