LCMV Beamforming via Two Parallel Linear Arrays

摘要

At high SNR, finite snapshots and the steering vector error may lead to serious degradation to the performance of LCMV beamforming. In the worst, a zero is set at the direction of the signal of interest and the signal is cancelled. In this paper, we start from the formula defining the weight vector of LCMV beamforming and point out that the mismatch between the signal model in the ill-conditioned correlation matrix and the apriori desired vector is the origin of the both problems. Based on this analysis, we put forward a method to do a non-orthogonal decomposition of the correlation matrix via the DOA matrix of two parallel linear arrays. The computer simulation shows the effectiveness of the method.