Research On User Scheduling In Millimeter Wave Hybrid Beamforming System

The fifth generation(5G)mobile communication system requires higher performance than the fourth generation(4G),and the use of the undeveloped spectrum in the millimeter-wave band is an effective means of supporting the high transmission rate of the 5G system.On the other hand,massive MIMO(Multiple-Input Multiple-Output,MIMO)technology combined with multi-user hybrid beamforming technology can greatly improve the system spectrum utilization,so massive MIMO technology has become an inevitable choice of 5G communication system.In an overloaded wireless communication system that uses a hybrid beamforming structure,a base station can serve multiple users in the same time-frequency resource to obtain multi-user multiplexing gains.However,in a practical system with multi-user equipping large-scale antenna array,a base station often cannot serve all candidate users on the same time-frequency resource.The inter-user interference always exists and the number of base station's radio frequency links is smaller than the number of candidate users.Therefore,a perfect user scheduling scheme will play a key role in obtaining good communication system performance.Different from the traditional user scheduling,which only needs to choose the best subset of users,the user scheduling needs to be solved in two dimensions in the mmwave hybrid beamforming system:the joint selection of subsets of users served by the base station on the same time-frequency resource and the communication beam pairs used by each selected user.Therefore,it has a more complicated processing flow.At present,the related research work is limited,and the algorithm has high complexity and needs to be based on the known channel status information(CSI).In view of the above problems,the research goal of this paper is to find a practical solution to the user scheduling problem in the millimeter-wave hybrid beamforming system,to find low complexity and work under unknown CSI conditions.First of all,aiming at the problem of high complexity of existing algorithms in known CSI scenarios,this paper proposes a heuristic user scheduling method based on user classification.The method simultaneously completes user selection and beam selection from the viewpoints of reducing the interference between served users and enhancing the energy of signals received by each served user.Computer simulation results show that this method can reach 92.8%of the optimal spectral efficiency under the low algorithm complexity.Then,considering the difficulty of obtaining complete CSI between users and base station in large-scale antenna system,this paper further proposes a user scheduling method based on beam training.The method can achieve 93.6%of the optimal system performance under the scenario that the user and the communication beam pairs are selected based on the beam training and can work in the scenario where all the users don't have known CSI,which is of high practical value and heuristic significance.