Network Sum-Rate Maximization for NOMA-Based Multicast Cognitive Radio Networks with SWIPT-Enabled Relays

In this paper, the problem of joint relay selection, power-splitting, and power allocation (J-RS-PS-PA) for network sum-rate maximization in non-orthogonal multiple-access (NOMA)-based multicast cognitive radio networks is considered. Specifically, the primary and secondary transmitters communicate - via simultaneous wireless information and power transfer (SWIPT)-enabled relays - with their respective receivers, subject to minimum rate requirements. Due to the non-convexity (and hence the excessive computational-complexity) of the formulated problem, it is decoupled and solved via a low-complexity solution procedure over two stages: (1) J-PS-PA, and (2) RS. For the first stage, an iterative two-layer algorithm is devised to maximize the sum-rate over each relay, while in the second stage, the optimal relay is selected. Simulation results revealed that the proposed solution procedure efficiently yields near-optimal network sum-rate as well as being superior to other benchmark schemes, while satisfying the minimum rate requirements of all primary and secondary receivers.