Computer Science · Signal Processing
Robust Adaptive Beamforming with Sidelobe Controlled
Peng Li, Ying Chen, Renhong Xie, Tianle Zhang, Xiaoqi Yuan, Yibin Rui, Shanhong Guo
See on the mapAbstract
High sidelobe level and model mismatches, such as general steering vector (SV) mismatches, are two major disadvantages of the minimum variance distortionless response (MVDR) beamforming. To suppress the sidelobe for interference nulling when encountering a mismatch problem, we propose a novel robust adaptive beamforming (RAB) technique with controlled sidelobe level. The proposed beamformer represents a modified MVDR that adds several new constraints to the original MVDR model for controlling the sidelobe. This approach estimates the desired signal SV by using small size and previously obtained imprecise array knowledge to maximize the beamformer output power. We get the final adaptive weight vector by solving the modified MVDR model with the estimated SV. The problem is shown to be convex, and thus can be solved using the CVX toolbox. Simulation results show that the proposed method has a significant performance improvement in output signal-to-interference-plus-noise ratio (SINR) and the controlled sidelobe.