Research On Cluster-based Wireless Resource Allocation In Ultra Dense Networks

The ultra dense network(UDN)is one of the key technologies of the fifth generation mobile communication system(5G).It can extremely improve network throughput and user experience rate by deploying small cell base stations densely.It can achieve seamless coverage of the whole network as well.However,with the increasing number of small cell base stations in the UDN,the network topology becomes more complex and it is harder to manage base stations.The interference and the energy consumption in the network becomes more and more serious as well.How to effectively mitigate interference,allocate network resources rationally,and then improve network throughput and energy efficiency are the main contents of this thesis.The main contents and achievements are as follows:(1)The technology of clustering in the UDN is studied.Aiming at solving the problem of complex network topology due to dense and random deployment of small cell base stations,a modified K-means clustering algorithm according to the distribution density of small cell base stations is proposed.The proposed algorithm divides base stations into different disjoint clusters dynamically according to the distribution density of small cell base stations,and realizes effective management of base stations and simplification of the network topology.(2)A cluster based network resource allocation scheme is studied.In the UDN,the interference problem caused by the dense deployment of the small cell base stations becomes the main factor to limit the system throughput.Therefore,a cluster based resource allocation scheme is proposed to mitigate interference by dividing mutually interfering base stations into different clusters,and then allocating orthogonal resource blocks(RBs)to users by adopting the technology of orthogonal frequency division multiplexing(OFDM).The RBs can be reused among the different clusters.And then,an improved two stage resource allocation algorithm based on greedy algorithm,which maximizes network throughput while considering user fairness and user's quality of service(QoS)is proposed.(3)A network resource allocation scheme based on overlapping clusters is studied.The intra-cluster interference can be solved by assigning the orthogonal RBs within the cluster,but the inter-cluster interference is still not solved.To solve this problem,the concept of overlapping clustersis proposed,which allows two adjacent clusters to overlap partially.The overlapping clusters allow a base station,which is near to the cluster edge,to join into another cluster to form an overlapping cluster.In the overlapping cluster,the problem of inter-cluster interference caused by the same reused resource block is solved by the technology of time division multiple access(TDMA).At the same time,the problem of network energy efficiency optimization is studied and the problem is transfered into a sub-problem of RBs allocation and a sub-problem of power allocation.Then,a distributed iterative power allocation algorithm based on non-cooperative game theory is proposed.The simulation results show that the proposed algorithm can effectively improve the energy efficiency of the network.