A Robust Adaptive Beamforming Algorithm Using Decomposition and Iterative Second-order Cone Programming

To overcome effectively the influence of large steering vector mismatch on the performance of adaptive beamformer, a Robust Adaptive Beamformer using Decomposition and Iterative Second-Order Cone Programming via Worst-Case performance optimization (RAB-DISOCP-WC) is proposed in this paper. Due to the decomposition and iterative method for the non-convex magnitude response constraints, the problem can be optimally solved using iterative Second-Order Cone Programming (SOCP), then the beamwidth and ripple of the robust response region can be flexibly controlled by the proposed method, and the output Signal-to- Interference-and-Noise Ratio (SINR) can be obviously improved. Moreover, in constrast to most of this class of robust beamformers, the proposed approach can get the optimal weight vector directly, and it does not need any spectral factorization. Thus, the proposed approach does not have any array geometry constraint, and it is applicable to arbitrary array geometries. Simulation results verify the correctness and validity of the proposed approach.