This paper investigates the cooperative full-duplex device-to-device (D2D) communication underlaying a cellular network, where the cellular user (CU) acts as a full-duplex relay to assist the D2D communication. To simultaneously support D2D relaying and uplink transmission, superposition coding and successive interference cancellation are adopted at the CU and the D2D receiver, respectively.The achievable rate region and joint outage probability are derived to characterize the performance of the considered system. An optimal power allocation scheme is proposed to maximize the minimum achievable rate. Besides, by analyzing the upper bound of the joint outage probability, we study a suboptimal power allocation to improve the outage performance. The simulation results confirm the theoretical analysis and the advantages of the proposed power allocation schemes.
Index TermsFull-duplex communication, D2D, superposition coding, successive interference cancellation, power allocation.self-interference (RSI) [15] caused by imperfect SIS. The feasibility and superiority of FD communication in non-cooperative D2D networks has been demonstrated in [16], [17], [18].In [19], the FD DT cooperates with the BS to perform non-orthogonal multiple access (NOMA)[20] and improves the outage performance of the user with weak cellular downlink. An adaptive multiple access switching method is proposed to dynamically choose the optimal multiple access scheme. As a dual-hop version of the model in [19], Zhang et al. investigate the optimal power allocation to minimize the outage probability [21]. To address the fairness issue between the NOMA-strong and NOMA-weak user, another power allocation scheme is studied to maximize the minimum rate achieved by the cellular and D2D link. Reference [22] considers the same cooperative scheme as in [5] with the DT operating in the FD mode. The cellular and D2D data are superposed in different power levels at the DT. Under the aggregate power constraint, an optimal power allocation algorithm is designed to maximize the achievable rates for the D2D users while fulfilling the minimum rate requirements of the cellular users. However, there is no SIC employed at the receiver of the cellular user and DR to deal with the mutual interference, which can be unsubstantial in some circumstances. The same model with amplify-and-forward relaying is discussed in [23]. A D2D based multicast service is considered in [24], one FD user equipment (UE) helps the BS convey data to a group of UEs. The FD D2D based multicast protocol has higher power efficiency than existing schemes, but the group size is limited to two UEs.
B. Motivation and Related WorkAll the works in [19], [21], [22], [23], [24] consider that there is always a direct link between DT and DR. When the DT and DR are separated far away from each other or the D2D link has poor quality, the D2D users either abandon the transmission or resort to the BS for data relaying [4], which limits the advantage of the D2D communication. To this end, relay-aided D2D communi...