Sum-rate maximization guaranteeing user fairness for NOMA in fading channels

Recently, non-orthogonal multiple access (NOMA) transmission has aroused an upsurge of interest due to its obvious superiority in spectral efficiency and user connectivity for the next generation cellular networks. However, as NOMA is intrinsically in favour of the users with strong channels who are capable of carrying out successive decoding, judicious design is required for ensuring user fairness. In this paper, we consider a two-user downlink NOMA with delay-tolerant transmission over fading channels in both scenarios of full and partial channel state information at the transmitter (CSIT). The average sum-rate is maximized subject to both an average and a peak power constraint as well as a minimum individual rate constraint. The dynamic resource allocation policy is optimally obtained using Lagrangian dual decomposition in the full CSIT case, while the power allocation in the partial CSIT case is also developed based on analytical results. Finally, the effectiveness of the proposed algorithms for NOMA over orthogonal multiple access (OMA) are verified in simulations by means of trade-offs for the average sum-rate and/or individual rate versus the minimum average rate requirement.