Optimal Resource Allocation in NOMA-assisted D2D Communication with Imperfect Channel State

Abstract

Device-to-device communication (D2D) is a promising technology of fifth-generation (5G) networks to maximize spectral efficiency. It permits direct communication between the mobile devices if they are in proximity by ignoring the base station. Despite improving spectral efficiency, it also improves the system data rate, minimizes the communication latency, and reduces the burden on the base station (for data uplink and downlink). The existing communication majorly focused on orthogonal multiple access (OMA), which serves one user at a time for spectrum sharing. To improve the spectral efficiency, non-orthogonal multiple access (NOMA) schemes can be employed that serve multiple users for spectrum sharing. NOMA drastically improves the spectrum sharing, but the devices can cause more interference that degrades overall the network performance. But, successive interference cancellation (SIC) of NOMA mitigates the interference issues and improves the system’s signal to interference noise ratio (SINR). Further, the system’s sum rate and secrecy capacity can be improved with a matching-based algorithm for optimal power allocation. This improves the overall sum rate, which further improves the system’s secrecy capacity against the number of eavesdroppers. Finally, the performance, i.e., overall sum rate and average secrecy capacity of the proposed system is evaluated by considering varying numbers of cellular users, D2D groups, and eavesdroppers.