Research On Power Control Technique For Uplink Non-Orthogonal Multiple Access

With the rapid development of mobile internet and internet of things,new industries and new services raise new challenges to the fifth generation(5G)wireless communications to support high throughput,low latency,high reliability,and massive connectivity.In order to address these requirements,non-orthogonal multiple access(NOMA)has been proposed as a potential alternative solution.Power-domain NOMA is one of concrete implementation schemes of NOMA.In the uplink transmission of power-domain NOMA,users will be divided into some clusters and the channel resources are shared by users in the same cluster.The transmit power of user is determined by power control,and the overlapped signals will be decoded by successive interference cancellation(SIC)receiver.User clustering and power control technique both have a great impact on the performance of power domain NOMA.Therefore,this thesis studies user clustering algorithm and power control strategy in uplink power-domain NOMA.Firstly,this thesis brings forward a novel user clustering algorithm,which is called interval clustering algorithm,and then this thesis explains the algorithm theory in detail.The proposed algorithm is compared with other algorithms,i.e.,greedy pairing algorithm,suboptimal clustering algorithm and hybrid pairing algorithm,in terms of performance of sum capacity.Simulation results indicate that the proposed algorithm can achieve better performance gains of sum capacity when multiple users have similar channel conditions.In addition,the proposed algorithm features no need of sorting of channel gains,dynamic scalability,better clustering flexibility.Secondly,this thesis proposes a non-linear power control strategy considering the nonlinearity of SIC receiver and fairness issue in uplink power-domain NOMA.With the assumption of delay channel quality,signal to interference ratio(SIR)for every user is approximately regarded as the same fixed value in the proposed strategy.The closed-form expressions of outage probability of users and sum capacity are obtained.Furthermore,the performance of sum capacity of NOMA and that of traditional OMA are compared when the fairness is guaranteed.Simulation results demonstrate that NOMA based on the proposed strategy can improve the sum capacity to a certain degree.Finally,based on the proposed non-linear power control strategy,this thesis evaluates the performance of SIC receiver with different number users in a cluster in uplink transmission of power-domain NOMA,and provides the cluster thresholds under some specified parameters.Evaluation results show that cluster threshold needs to be adjusted reasonably according to different requirements of system designing and actual scenario.