Eigenspace-based linearly constrained minimum variance beamformer

The eigenspace-based linearly constrained minimum variance beamformer (ELCMVB) is presented; it combines the linearly constrained minimum variance beamformer (LCMVB) with the eigenspace-based beamformer. The ELCMVB projects the presumed steering vector of the LCMVB onto the signal subspace, the projected steering vector is then used to calculate the weight vector for beamforming with the linearly constrained minimum variance technique. Compared to the generalized eigenspace-based beamformer (GEIB), the ELCMVB removes the computation of the modified signal subspace. It can thus avoid numerical instability. The theoretical analysis also indicates that the ELCMVB performance is not affected by the positions of the constraints. Computer simulation results are presented and demonstrate the merits of the ELCMVB.