Antennas' correlation influence on the GMD-assisted MIMO channels performance

Abstract

The use of multiple antennas in MIMO (multiple-input multiple-output) systems at both the transmit and receive sides produces the effect known as antennas correlation which impact the overall channel performance, throughput and bit-error rate (BER). The geometric mean decomposition (GMD) is a signal processing technique which can be used to process transmit and receive signals in MIMO channels. The GMD pre- and post-procesing in conjunction with dirty-paper precoding shows some advantages over the popular singular value decomposition (SVD) technique which provides GMD-assisted MIMO systems a superior performance particularly when the channel is affected by antennas correlation. This paper analyses the impact of antennas correlation on the performance of GMD-assisted wireless MIMO channels highlighting the advantages over SVD-assisted ones.