Research On User Association And Cooperation Of Multiple Points Scheduling Algorithm In Heterogeneous Networks

Heterogeneous network is one of the most important networking candidates for the coming 5G mobile communications systems.It has the following advantages,including more flexible network deployment,better coverage ability,and higher energy/spectrum utilization,which is helpful to 5G system to achieve the version of “information a finger away,everything in touch”.However,due to the deployment of low-power nodes,the transmission power difference between Macro stations and Pico stations,and inter-cell interference introduces several new challenges,especially for network deployment,user assiciation and resource scheduling.To address these challenges,this disseration investigate the problem of user assiciation and resource allocation for Het Nets,aiming at maxmizing the throughput of cell-edge users with considerations of load balacing and the throughput of overall system.The main contents and innovations on this paper are summarized as follows:For user association,this paper proposes a user association algorithm named GA-BP algorithm based on Greedy Algorithm and Back Propagation neural network by studying the existing user association algorithm.The core idea of GA-BP is to obtain the parameters and results of the user association based on greedy algorithm,and considering these data as the training data.Then,the training BP network can be got by its self-learning.Finally,through the trained BP neural network,user association results can be accurately predicted and judged.The simulation results demonstrate that the proposed algorithm can greatly reduce the dec ision time while guaranteeing good user association results.Also,this algorithm can get the selection result of cooperative base station cluster.As for resource scheduling,the allocation algorithm of Resource Block(RB)under heterogeneous network is studied based on the research of existed user association algorithms.In this paper,a binary particle swarm optimization algorithm based on distributed and centralized Combination(BPSO-DCC)is proposed based on the analysis of the RB resource allocation process in the Co MP system.The core idea of BPSO-DCC is to decompose the whole scheduling process into 3 stages: RB division,distributed and centralized cooperative scheduling and scheduling based on binary PSO.The simulation results demonstrate that the algorithm can effectively reduce the competition and the computational complexity during RB allocation process,improve the edge user throughput,as well as maximize the system performance by allocating RBs reasonly for the edge users and the central users.