Robust beamforming algorithm based on steering vector estimation and interference plus noise covariance matrix reconstruction

We are focusing on the problem that serious performance degradation of general adaptive beamformers in the presence of model mismatch error. In this paper, a robust beamforming algorithm based on steering vector (SV) estimation and interference plus noise covariance matrix (INCM) reconstruction is proposed. First, the estimated value of target SV is obtained by sparse reconstruction and the INCM is reconstructed by eliminating the expected signal components in the sample covariance matrix. Then, based on the knowledge of subspace expansion the SV is optimized by establishing constraint model, after that INCM is reconstructed based on the relationship between the signal SV and the INCM. Finally, the weight vector is obtained with optimal SV and INCM, and the optimal value is obtained by iteration. Simulation results show that the algorithm improves the performance of the beamformer in the case of desired signal steering vector error and sensor location error.