Optimal Source and Relay Design for Multiuser MIMO AF Relay Communication Systems With Direct Links and Imperfect Channel Information

Abstract— We propose statistically robust design for multiuser multiple-input multiple-output (MIMO) relay systems with direct source-destination links and imperfect channel state information ( CSI). T he minimum mean-squared error (MMSE) of the signal waveform estimation at the destination node is adopted as the design criterion. We develop two iterative methods to solve the nonconvex joint source, relay, and receiver optimization problem. Simulation results demonstrate the improved robustness of the proposed algorithms against CSI errors. The optimal relay precoding matrix maximizing the source-destination capacity has been developed for AF MIMO relay systems. However, in real communication systems there is always mismatch between the true and the estimated CSI, due to channel noise, quantization errors, and outdated channel estimates. A unified framework for joint source and relay matrices optimizing in linear AF multicarrier. The joint source, relay, and receiver matrices design for multiuser MIMO relay systems with direct source-destination links and CSI mismatch. The effect of CSI mismatch is shown in the structure of the optimal robust source and relay matrices. The destination node obtains all CSI required. Then the destination node performs the transceiver optimization and sends the optimized F1i. < final year projects >