Research On Resource Allocation And User Association Of Dense Heterogeneous Networks Deployments

With the wide and commercial use of 4G communication system,an irresistible wave of momentum of the 5G era is coming to us.Dense deployment of heterogeneous networks can shorten the access distance of communication nodes,therefore,it has become a hotspot in current cellular communication academic research.As a vital member of the key technology of 5G,it greatly improves coverage range and service quality compared with that of the original cellular network,which works only depending on macro stations or a few heterogeneous small cells.However,there are still some problems at the application stage,such as the division of scarce spectrum resources,user association problem and power consumption caused by a large number of small base stations.On the basis of the above introduction,the resource allocation and user association problems in dense heterogeneous networks deployments are studied in this thesis,which trying to improve the performance of existing algorithms.Firstly,as dense deployment of heterogeneous networks is still in research and discussion,there is no agreement on the definition of this concept.Therefore,by collecting and summarizing related literatures,this thesis classifies them based on the relationship between deployment density of base stations and users.In addition,potential research directions that are suitable for each type of the definition are also analyzed.Specifically,when the base station deployment density is much larger than that of users,resource allocation problem is comparatively outstanding.On the contrary,when there is no restriction on the relationship between absolute density of them,user association is a relatively more important problem to be solved.Secondly,based on the original potential game algorithm,a novel bidirectional interference graph-based resource allocation algorithm is proposed,which improvs the performance of system throughput and reduces computational complexity during the iterative game process in shared spectrum access under the guarantee of QoS of cellular users.Additionally,the throughput of edge users in dedicated spectrum access is effectively advanced based on the improved graph coloring label-based algorithm in this thesis.Finally,a mixed max-SINR user association algorithm which uses matching theory and modified expression of power consumption optimization in the current literature is proposed in this thesis.Simulation results show that,compared with the original algorithm,the proposed algorithm both increases the energy efficiency and fairness during user association process.