Research On Power Allocation And User Grouping Algorithms For Non-orthogonal Multiple Access System

Multiple access has always played a key role in the evolution of mobile communication systems.Traditional frequency division multiple access,time division multiple access,code division multiple access and orthogonal frequency division multiple access mainly use orthogonal resource allocation.However,orthogonal resource allocation has the disadvantages of limited spectrum efficiency and complex time-frequency resource scheduling.It has been difficult to meet the requirements of high-capacity and high-spectrum efficiency of future mobile communication systems.New multiple access realizes the non-orthogonal superposition of multi-user signals on physical resources by introducing different resource dimensions,which can not only further increase spectrum efficiency,but also significantly increase system access capacity.This thesis focuses on the power allocation and user grouping scheme in power domain non-orthogonal multiple access(NOMA)system.The main work is as follows:Firstly,the basic principle of the NOMA system is described in detail,and with the two multiplexing users as example,the advantages of the NOMA system in spectrum efficiency are further elaborated.Based on the downlink NOMA system model,the principles of the several common transmitter power allocation algorithms,user grouping algorithms,and interference cancellation algorithms at receiver are introduced.Secondly,power allocation for the multiplexing users in the same subband and the grouping of cell users are studied.The step-size adjustable power allocation algorithm and multi-user grouping optimization algorithm are proposed respectively.The proposed step-size adjustable power allocation algorithm is aimed at weighted throughput sum maximization for the multiplexing users in the same subband,and the increasing power allocation coefficient satisfying the SIC requirement is a constraint condition.The range of power allocation coefficient search is controlled flexibly by adjustable step-size.Meanwhile the redundant power allocation coefficient combinations are identified and discarded step by step in the power allocation coefficient search process according to the weighted throughput sum maximum standard.The proposed power allocation algorithm can achieve a compromise of system throughput,computational complexity and user fairness.According to the user channelgain difference and the restrictions of the multiplexed user number in the same subband,the proposed multi-user grouping algorithm firstly performs the process of the initial multi-user grouping to reduce the user's search space.Then,the optimized combination of the initial grouping users are gradually completed,and the maximum geometric mean user throughput is used as user grouping criterion,which can further enhance the cell-edge user throughput.Finally,a power allocation algorithm based on energy efficiency optimization is proposed for the energy efficiency optimization problem of the multiplexing users in the same subband.The proposed algorithm fully considers the requirements of quality of service and fairness for multiplexing users in the same subband.Under the premise of satisfying the subband minimum transmission power,an energy efficiency optimization model for multiplexing users in the same subband is given.Then,the closed-form expression of the power allocation coefficients are deduced by KKT conditions,and the power consumption ratio is solved by the fractional programming Dinkelbach algorithm.The simulation results show that the proposed algorithm has better energy efficiency than the traditional spectral efficiency maximizing power allocation algorithm under the premise of meeting the requirements of quality of service and fairness for the multiplexing users in the same subband.