Research On Resource Allocation Algorithm In Downlink NOMA

In the NOMA technology,multiple users share the same time-frequency domain resource for transmission,because of its high spectral efficiency,it is favored in the research of 5G technology.and the superimposed users are distinguished according to power,therefore,user grouping and power allocation algorithm will greatly affect the performance of the system.At present,there are still some deficiencies in the research of NOMA user grouping and power allocation algorithm,the user channel gain is the main consideration in the study of user grouping algorithm,there are problems such as user fairness is not guaranteed and complexity is too high.In addition,in the research of power allocation,it is generally the optimization of the system capacity for the superposition of two users without considering the channel estimation error,without considering the multi-user superposition and the quality of service of the users.For the problem that the complexity of user grouping in the downlink of NOMA is too high and the data rate of edge users is not guaranteed,this thesis converts the user grouping problem into optimal matching problems for weighted bipartite graphs,the Kuhu-Munkres algorithm in the optimal matching theory is used for user grouping,and proportional fair algorithm is used for power allocation among users of the same group of users.The simulation results show that the proposed algorithm not only guarantees the data rate of edge users but also has better system capacity,and because of the optimal matching algorithm,the computational complexity is low.For the problem that power allocation does not consider the number of superimposed users and channel estimation error,under the condition that the channel state information is not fully known in this thesis,each user is guaranteed to meet certain quality of service,and the system capacity is maximized as the optimization goal,in this thesis,a step-by-step power allocation algorithm is proposed.In the algorithm,the complex Gaussian distribution is used to represent the channel estimation error,and the minimum data rate threshold of each user data rate under this channel condition is introduced into the power optimization objective to ensure the quality of service of the users.Considering the complexity of the optimization problem solving,this thesis decomposes the power allocation problem into two sub-problems: inter-band power allocation and in-band power allocation.Firstly,the linear water-filling algorithm is used to complete the inter-band power allocation,and then the Karush-Kuhn-Tucker optimal constraint is applied to maximize the system capacity.Simulation results show that the proposed algorithm has higher system capacity and lower computational complexity than the existing power allocation algorithm.