Multiuser AF MIMO Multi-Relay System Design with Direct Links and MMSE-DFE Receiver

Adopting the amplify-and-forward (AF) relay protocol, this paper focuses on the design of a multiuser multiple-input multiple-output (MIMO) multi-relay system with direct source-destination links taken into consideration. In order to improve the quality of signal detection at the receiver, we make use of the minimal mean-squared error (MMSE)-decision feedback equalization (DFE) technique. With the constraints of the signal transmission power at both source and relay nodes, the minimization of the sum mean-squared error (MSE) for the signal waveform estimation of all users' data streams is employed as our design criterion. Through the block coordinate descent (BCD) method of Gauss-Seidel type, we develop an iterative algorithm with guaranteed convergence to conduct the joint optimization of all the source precoding, relay amplifying, feed-forward and decision feedback matrices, where each step we take is to solve a convex problem. It is shown from simulation results that, in comparison to the linear MMSE receiver-based algorithm, the proposed nonlinear one achieves better performance in terms of both MSE and bit-error-rate (BER). Moreover, the system reliability can be significantly improved when the signal-to-noise ratios (SNRs) of direct links become relatively good.