This study analyses the performance of a non-orthogonal multiple access-based two-way relaying system. Analytical expressions are obtained to characterise the outage probability and outage floor of the system for end-to-end communication over Rayleigh fading channels. Furthermore, results are also presented for the resulting individual user and sum ergodic capacities. A user power allocation technique is proposed for sum-rate maximisation with user quality of service constraints by transforming the original non-convex problem into a difference of convex program that can be readily solved. Finally, simulation results are presented to authenticate the analytical expressions obtained for the various quantities and illustrate the efficacy of the proposed scheme.

Non-orthogonal multiple access (NOMA) has recently attracted a lot of attention as a promising multiple access technology to be used in the fifth generation (5G) cellular networks. In this study, the authors investigate uplink NOMA transmissions in a cooperative cellular network in the presence of frequency selective multipath fading channels. Specifically, two users communicate with the base station simultaneously with existence of a half-duplex relay employing the decode-and-forward scheme to assist the far user. At the transmitting nodes, the power allocation factors are decided based on the statistical channel state information (CSI) rather than perfect CSI, while at the receiving nodes, the decoding order between the users is pre-specified and the effect of imperfect successive interference cancellation implementation is taken into account. Under Rayleigh fading channels, the authors derive the exact ergodic rate for the near user and the upper bound of the ergodic rate for the far user, and present an ad-hoc approach to determine the power allocation factors when the target data rates are given. Simulation results show that NOMA consistently outperforms orthogonal multiple access, however, the performance of uplink NOMA depends heavily on the proper choices of the power allocation factors and the relay location.

In this study, the authors analyse the average bit-error rate (BER) of a wireless powered three-node decode-and-forward (DF) relay system in a Rayleigh fading environment. The direct link between the source and the destination is present and the received signals are combined using equal gain combining at the destination. Data is assumed to be binary phase-shift keying modulated. The relay node of the system is powered by the source node, i.e. the relay node harvests energy using radio frequency transmitted at the source node and forwards the signal using the harvested energy. The source data is communicated in three time slots; energy harvesting in the first time slot and selective DF relaying in the second and third time slots. The asymptotic expression of the average BER is analysed and used for the suboptimal energy allocation in different time slots. Since energy consumed in a slot depends on the transmitted power and the slot duration, the relation between them is used for the suboptimal allocation of the transmit power and time among different slots.

Joint beamforming optimisation for a wireless-powered multi-pair two-way relay network (MP-TWRN) is studied. Different from the existing works, the authors investigate relay beamforming and power-domain non-orthogonal multiple access for the wireless powered MP-TWRN. The authors' goal is to optimise the energy-transfer beamforming matrix and the relaying beamforming matrix in the sense of maximising the minimum of the achievable rates among all the users with the power constraint at the relay. Further, they explore the above issue under two relaying schemes: amplify-and-forward and decode-and-forward. The formulated problems are highly non-convex, and thus, it is challenging to solve. To address these, they propose two iterative algorithms to solve those two problems based on the difference of convex programming transformation and the successive convex approximation. Simulations demonstrate that the proposed schemes outperform the orthogonal multiple access schemes.

Recently, non-orthogonal multiple access (NOMA) has been considered as a promising radio access technique for the upcoming fifth generation networks. In this study, the authors consider the downlink multi-cell multi-user cellular network, where the base station (BS) located in the centre of each cell intends to communicate with multiple random distributed user equipments (UEs) using NOMA protocol. They formulate an optimisation problem to increase the sum network capacity under various practical constraints. Especially, the transmission power budgets for BSs, power allocation for UEs, and minimum rate requirements per UE are optimised. Moreover, they obtain the optimal power allocation via local optimal solution iteratively, which satisfies Karush–Kuhn–Tucker conditions. The simulation results show that the proposed scheme can converge after few iterations and achieve higher sum capacity compared with the non-optimal NOMA and orthogonal multiple access schemes.