Relay and antenna selection in multi-antenna amplify-and-forward (AF) systems with partial channel state information

This paper combines the idea of relay and antenna selection with that of partial channel state information (CSI) based precoding. We consider a non-regenerative two-hop relay network where all the nodes are equipped with multiple antennas. It is assumed that multiple relay nodes are available for cooperation and that the source and each individual relay have the full CSI of the source-relay channel. It is further assumed that each relay knows the spatial correlation matrix of relay-destination channel. The structure of the optimum precoders to minimize a tight lower-bound on the average received minimum mean square error (MMSE) at the source and the relays are then introduced. Using these precoders and based on a computationally efficient proposed method, the best relay and antenna subsets at the source and the relay are then selected to further minimize the MMSE. By simulation, we show that the proposed scheme provides a significant performance gain in terms of the bit error rate (BER).