Iterative Diagonally Loaded Sample Matrix Inverse Robust Adaptive Beamforming

To overcome effectively the influence of large steering vector mismatch on the performance of adaptive beamformer, an Iterative diagonally Loaded Sample Matrix Inverse (ILSMI) robust adaptive beamformer is proposed in this paper. The iterative computation of conventional diagonally Loaded Sample Matrix Inverse (LSMI) method is implemented in the proposed approach. Based on the relationship between the optimal weight vector and the assumed steering vector of the Capon beamformer, the more precise corresponding steering vector of the LSMI optimal weight vector is estimated in each iteration and used to replace the assumed steering vector, and the converged estimation will approach to the real steering vector. The proposed approach can obviously improve the output Signal-to-Interference-and-Noise Ratio (SINR) of the robust beamformer through only one key recursive formula, without any Lagrange multiplier methodology or convex optimization methods in each iteration. The simulation results verify the correctness and validity of the proposed approach.