Research On System Scheduling And Resource Allocation For NOMA

The new non-orthogonal multiple access(NOMA)technology performs multiplexing by introducing new dimensions in the time-,frequency-or code domain.Under the background of 5G explosive data growth,improving the spectral efficiency and energy efficiency of NOMA have become the research focus.Traditional resource allocation schemes can't meet it's needs,so this thesis focuses on user pairing and power allocation scheme in power domain NOMA systems.The main work is as follows:Firstly,based on the scenario of two multiplexed users in the NOMA downlink systems,the basic principle of the NOMA systems and the technical advantages compared with the traditional orthogonal multiple access systems are described.Simultaneously,several common receiving and detecting technologies,user pairing and power allocation algorithms are analysised.Secondly,for the user pairing and resource allocation optimization problems of the capacity in NOMA systems,an extension of the work is converting the double-objective optimization problem into a single-objective optimization problem by power discretization method.The global optimal search algorithm is proposed to determine the upper bound of system throughput.In order to achieve a compromise among system throughput,number of accessed users and users' fairness,a low complexity algorithm with step-by-step optimization is proposed.To improve the number of accessed users,an adaptive proportional fair user pairing scheme is designed,which dynamically adjusts the forgetting factor based on the variance of all the users' scheduling priorities,so as to influence the update of users' scheduling weights.At the same time,the power allocation scheme with proportional rate constraint is used to further improve the fairness of the paired users,it could obtain a relatively stable result while allocating power.The simulation results show that the proposed algorithm could effectively ensure the throughput of the NOMA systems while improving the number of accessed users and users' fairness.Finally,the originally non-convex optimization problem of the energy efficiency user pairing and resource allocation in NOMA systems is decomposed into three optimization subproblems.In the user-subchannel matching phase,an adaptive proportional fair user pairing scheme based on equivalent channel gain is adopted.And the equal power allocation scheme is adopted across subchannels.Then the power allocation optimization problem between multiplexed users on each subchannel is converted into a DC programming problem,which can be solved by the concave convex procedure algorithm.Meanwhile,in order to guarantee the QoS of the paired users,a minimum power requirement scheme on each subchannel is derived.It's determined by the paired users' equivalent channel gain,the users' minimum data rate requirement and the upper bound of power allocation factor between multiplexed users on each subchannel.The simulation results show that the proposed algorithm offers higher system throughput and energy efficiency than the traditional energy efficiency resource allocation algorithm.