Joint Relay Clustering and Beamforming Design for Cooperative Relay Networks

Consider a multi-cluster cooperative relay network, where each relay cluster (consists of a certain number of amplify-and-forward (AF) relays) forwards the signal from its associated user equipment (UE) to the base station (BS). With the goal of providing fairness among the UEs and reducing the costs of full relay cooperation, we study the joint design of relay clustering and beamforming to maximize the minimum signal-to-interference-and-noise ratio (SINR) under per relay power constraints. This max-min SINR problem with relay clustering is formulated as a mixed-integer programming (MIP) problem, which is generally NP-hard. To tackle this problem, we propose a block coordinate descent (BCD) based algorithm based on the property that the constraints are separable among the optimization variables, i.e., the clustering matrix, the relay cooperative beamforming vector, and the receive beamforming vectors at the BS. Specifically, these variables are optimized iteratively in an alternating fashion, one at each time with others being fixed and we show that each subproblem can be efficiently and optimally solved. Simulation results demonstrate the effectiveness of the proposed algorithm as compared with the benchmark schemes.