Source Precoder Design for Non-Regenerative MIMO Relay Networks with Antenna Selection

In this paper, we present a source precoder design and relay antenna selection by exploiting the sparsity of the relay gain vector. We consider the Amplify-and-Forward (AF) protocol in a single relay scenario equipped with multiple spatially correlated antennas. We investigate the performance of the proposed model in practical scenarios, where the antennas are spatially correlated and the Channel State Information (CSI) is imperfect at all terminals. For the precoder design, we present a technique which diagonalizes the Mean Squared Error (MSE) matrix using singular value decomposition (SVD) and the generalized SVD (GSVD). The proposed solution for the precoder aims at minimizing the MSE and reducing the design complexity. As for the relay antenna selection scheme, we employ a technique based on the Orthogonal Matching Pursuit (OMP) algorithm to generate the antenna selection matrix. Extensive Monte Carlo simulation results demonstrate that the performance gain of the proposed schemes over the existing ones is significant.