Spectrum-Power Trading for Energy-Efficient Device-Centric Overlaying Communications

In this paper, we propose device-to-device (D2D) overlaying communications with spectrum-power trading where D2D users (DUs) consume transmit power to relay the data of cell-edge cellular users (CUs) for uplink transmission in exchange for bandwidth from CUs for D2D communications. The proposed spectrum-power trading aims at exploiting individual disparities from both the spectrum and the power perspectives. Our goal is to maximize the weighted sum EE (WSEE) of DUs via a joint D2D relay selection, bandwidth allocation, and power allocation while guaranteeing the quality of service of each CU. We show that for a given D2D relay selection, the objective function of the WSEE maximization problem in a fractional form can be transformed into a subtractive-form that is more tractable based on the fractional programming theory. To perform D2D relay selection, we first reveal an important property, which connects the WSEE with both the system-centric EE and the fairness- centric EE. Based on this insight, the D2D relay selection problem is cast into a minimum weighted bipartite matching problem that can be solved efficiently with optimality. Simulation results demonstrate the effectiveness of the proposed scheme and algorithm.