Adaptive Design for Colocated MIMO Radar Beamspace Localization

The angle estimation in the presence of interfering sources is an important problem for colocated MIMO radar. In this paper, we design the beamspace transformation matrix to adaptively filter out the interferences in the received signals by assuming that the angular sector covering the target directions is known. To tackle the nonconvex optimization problem encountered in the design, an iterative algorithm based on alternating direction method of multipliers is proposed. Compared with the existing data-adaptive beamspace design algorithm, the beampattern associated with the proposed algorithm achieves a smoother mainlobe and lower sidelobes. Moreover, the designed transformation matrix can maintain the spatial whiteness of the transformed noise, which is important for many angle estimation algorithms. Numerical examples show that the proposed algorithm achieves a higher localization accuracy in the low signal-to-noise ratio region.