Robust Adaptive Beamforming Based on Steering Vector Estimation and Covariance Matrix Reconstruction

Abstract

In this letter, a novel robust adaptive beamforming (RAB) technique is proposed. It offers robust performance even with large look direction error in the array steering vector (ASV) and an uncertainty in the covariance matrix. In essence, the proposed ASV estimation technique treats the ASV as a vector lying within the intersection of two subspaces, and then estimates it using a closed-form formula. In this technique, the covariance matrix is reconstructed after the desired signal (DS) eigenvalue is replaced by the average value of the noise eigenvalues, thus eliminating a large portion of the DS. Furthermore, the only prior information necessary is knowledge of the antenna array geometry and angular sector in which the actual ASV lies. Simulation results indicate that the proposed method achieves good performance in terms of output signal-to-interference-plus-noise ratio (SINR), as long as the input signal-to-noise ratio (SNR) is not close to the interference-to-noise ratio (INR) and the DS and interference signals are well separated.