Sum-Rate Optimization for Device-to-Device Communications over Rayleigh Fading Channel

In this paper, we investigate the sum-rate maximization in the Device-to-Device (D2D) communication underlaying cellular networks, where many cellular users (CUs) share the uplink resource with the D2D pairs. We show that the system sum- rate maximization problem can be formulated as a mixed integer non-linear programming problem, which is NP-hard in general. We circumvent this difficulty by applying the optimization decomposition: 1) Given a resource allocation policy, we derive the optimal Signal to Interference plus Noise Ratio (SINR) threshold to maximize the system sum-rate. 2) We then propose a coalition game approach to further optimize the resource allocation policy, and prove that the proposed coalition game approach can converge to the Nash-stable partition in finite time. Simulation results show that 1) the performance of the coalition game is close to the exhaustive search, but its run-time is much shorter than the exhaustive search; 2) compared with several other resource allocation policies, the coalition game can achieve an average sum-rate improvement of 13%-173%, and has the best resource sharing fairness.