MMSE-DFE Based MIMO Relay System with Correlated Fading Channel

We consider a practical dual-hop nonregenerative multiple-input multiple-output (MIMO) relay system, where the relay node only knows the correlation matrix of the relay-destination channel. A nonlinear minimal mean-squared error (MMSE)-based decision feedback equalizer (DFE) is used at the destination node to retrieve the source signals. We derive the structure of statistically optimal source and relay precoding matrices to minimize a class of objective functions which are multiplicatively Schur-convex with respect to the diagonal elements of the MSE matrix. Simulation results demonstrate that the proposed algorithm has a very close performance compared to MIMO relay system with full channel knowledge at the relay node, and thus is very useful for practical relay systems.