MSE based antenna selection for MIMO-SDM systems

Antenna selection is a promising technique which provides performance improvement for multiple input multiple output (MIMO) systems at low cost of additionally employed antennas and signal processing. Various antenna selection algorithms based on different criteria have been proposed in the literature. This paper presents a new antenna selection algorithm for MIMO spatial division multiplexing (MIMO-SDM) systems which uses the mean square error (MSE) as a criterion for selecting the best subset of transmit/receive antennas to improve the system error performance. The proposed algorithm helps to achieve lower the bit error rate (BER) for MIMO-SDM systems with linear detectors compared with the previous channel matrix norm based and signal to noise ratio (SNR) based selection algorithms. Compared with the SNR based algorithm, the proposed algorithm provides double diversity gain in receive antenna selection systems and the same diversity order but about 1 dB SNR gain in transmit antenna selection systems.