A Game-Theoretic Approach to Distributed Energy-Efficiency Maximization in Uplink NOMA Relay Ad-Hoc Networks

This paper considers distributed energy-efficiency maximization in uplink non-orthogonal multiple-access (NOMA) relay ad-hoc networks. Specifically, the aim is to allow network users to maximize their energy-efficiency (EE) in a distributed manner over an energy-harvesting amplify-and-forward relay, while satisfying their quality-of-service (QoS) constraints. The centralized energy-efficiency maximizing power allocation (C- EE-MAX-PA) problem is formulated as a multi-objective optimization problem, which is NP-complete (i.e. computationally- prohibitive). In turn, a distributed game-theoretic power control algorithm is proposed, whereby each user iteratively updates its transmit energy such that its EE is maximized, ultimately converging to the generalized Nash Equilibrium (GNE). Simulation results are presented to evaluate the proposed algorithm, which is shown to yield the Pareto-optimal solution of problem C- EE-MAX-PA; however, at lower computational complexity, while satisfying the users' QoS constraints.