Joint Source and Relay Precoding Designs for MIMO Two-Way Relay Systems

We investigate the source and relay precoding based on the minimum mean-square-error (MMSE) criterion for amplify-and-forward (AF) MIMO two-way relay systems. This joint design problem is shown to be a highly nonconvex optimization problem. In this work, we present two efficient design algorithms. The first one aims to minimize the total MSE of two users by alternatively solving three trackable sub-problems and it is iterative in nature. Since the optimal solution for each sub-problem can be obtained, the convergence is thus ensured. The second design aims to compromise computational complexity and system performance and possesses a certain precoder structure. This structure is able to parallelize channels in the Multiple Access (MAC) and Broadcast (BC) phases of the two-way relaying. Based on such structure, the joint precoding design is simply reduced to the joint source and relay power allocation problem. The efficiency of both proposed algorithms is verified through simulation.