Efficient resource allocation for OFDMA-based device-to-device communication underlaying cellular networks

In this paper, we investigate the resource allocation problem in OFDMA-based device-to-device (D2D) communication underlaying cellular networks, where we aim to maximize the sum rate of the cellular system. The formulated challenging optimization task is divided into two subproblems: D2D pair assignment and power distribution. We show that the D2D pair association falls into the maximum assignment problem class and develop an efficient time-sharing method to address it. A low complexity iterative algorithm is developed to deal with the power allocation problem, where the locally optimal solution is approached by solving a series of convex optimization problems that can yield a tight lower bound of the sum rate. The convergence of our proposed algorithm is proved and confirmed by numerical results.