MMSE Based Transceiver Design for MIMO Relay Systems with Mean and Covariance Feedback

In this paper, the problem of transceiver design in a non-regenerative MIMO relay system is addressed, where linear signal processing is applied at the source, relay and destination to minimize the mean-squared error (MSE) of the signal waveform estimation at the destination. In the proposed design scheme, optimal structure of the source and relay precoding matrices are obtained with the assumption that the relay knows the mean and channel covariance information (CCI) of the relay-destination link and the full channel state information (CSI) of the source-relay link. Based on this assumption, an iterative joint source and relay precoder design is proposed to achieve the minimum MSE of the signal estimation at the destination. In order to reduce computational complexity of the proposed iterative design, a suboptimal relay-only precoder design is proposed. A numerical example shows that the performance of the proposed iterative joint source and relay precoder design is very close to that of the algorithm using full CSI.