Joint User Association, Grouping and Power Allocation in Uplink NOMA Systems with QoS Constraints

Abstract

To alleviate severe inter-user interference in Non-orthogonal multiple access (NOMA) systems, in practice, several users are grouped together for NOMA transmission and orthogonal resources are allocated among different groups. In this paper, we investigate uplink NOMA scenarios with multiple base stations, where user association is jointly considered with user grouping. It is worth noting that a user prefers to be associated with the base station where it can be assigned into a group with less inter-user interference. Furthermore, different users have different quality-of-service (QoS) requirements. Based on these observations, we firstly quantitatively analyze the extra transmit power that a user brings to the others in the same group under QoS constraints, i.e., externalities, and derive an externality function to indicate interference among users in uplink NOMA. Then, we obtain a game model for the joint user association, grouping and power allocation (JAGP) problem in uplink NOMA systems and prove that it is an exact potential game, with the goal to minimize total transmit power. Furthermore, we prove the existence of Nash Equilibrium (NE) as well as the finite improvement property (FIP) of this game. In addition, the NE of this game is proved to be Pareto-optimal. A heuristic algorithm is proposed to find the solution to the JAGP problem by achieving the NE of the proposed game, which convergence is guaranteed by FIP. Simulation results show that the proposed algorithm can considerably reduce the total transmit power compared with existing schemes.