Adaptive clustering with continuous phase modulation in NOMA systems

Abstract

ABSTRACT Power-domain NOMA can enlarge the spectrum efficiency and capacity via serving multiple users in the same time slot. Up to now, there are still some issues to be addressed in the NOMA system with multiple users: how many users the NOMA system can accommodate with limited power, how to mitigate the interference among the users, what’s the achievable bound of capacity with limited power at the base station, and how to tackle with two or more users having similar distances to the base station. In the paper, the approximated CPM capacity in the NOMA system over Rayleigh channel is derived. Then, an adaptive user-clustering is proposed in downlink NOMA system, such that the achievable sum-rate will get high enough. Moreover, the optimal allocation of power for each user in each cluster is derived and evaluated. In the end, the performances such as outage probability, capacity, peak-to-average power ratio (PAPR), and bit error ratio (BER) are simulated as comparison.