Design of MIMO communication systems robust to eigenvector mismatch problem

Abstract

This work presents a design method of multiple input multiple output (MIMO) wireless systems that shows robustness to eigenvector mismatch problem between the transmitter and receiver arrays. This problem often happens when the first and second largest eigenvalues are close together that results in the disconnection or quality degradation of communications. The proposed method is based on the recovery of the correct weight pair between the transceiver and receiver. First, to cope with the uncertainness of eigenvectors coming from the eigenvalue degeneration of the channel correlation matrix, we give a procedure to derive a unique weight common in both arrays. Then through computer simulations, it is shown that this approach achieves high SNR without losing the orthogonality of weights for plural MIMO paths as in the case of the eigenvector exchange when the relation of the eigenvalues is changed in one side of the arrays.