Joint User Pairing and Resource Allocation for Buffer-aided NOMA Networks

Abstract

Due to the advantages of improving network connectivity and spectral efficiency, non-orthogonal multiple access (NOMA) has been regarded as a promising technology for future networks. Meanwhile, buffer-aided communication technology, can greatly enhance network throughput via adaptive traffic control and transmission scheduling, and has attracted many research attentions. In this paper, we study the achieved long-term network throughput of a buffer-aided NOMA network, wherein the data buffer stability is taken into account, and the data admission control, power allocation and user pairing policies are jointly optimized. To solve the formulated network throughput maximization problem, we use the Lyapunov optimization framework to decouple the long-term time-averaged throughput maximization problem, into multiple single-slot resource allocation and user pairing optimization problem. On the basis, we propose an online scheme which includes data control, power allocation and user pairing. Specifically, the close-formed power allocation scheme is derived with given number of NOMA users by the decomposition method. Furthermore, we design a user pairing scheme to construct the user set for NOMA transmission in the same communication channels. Finally, extensive simulation results validate that the proposed adaptive resource scheduling algorithm greatly improves network throughput while strictly maintaining buffer stability.