Research On Interference Management Algorithm For Ultra-Dense Networks

In order to meet the needs of super-strong broadband connection,ultra-low delay communication and ultra-reliable secure communication,the Fifth-generation Mobile Communication System(5G)has introduced many key technologies.As one of the key technologies of 5G,ultra-dense network is a hot technology for wireless access.In ultra-dense networks,with the massive deployment of low-power small base stations,the quality of communication links and network capacity of users continue to improve,while the interference among cells becomes more and more serious,which not only reduces users’ communication experience,but also seriously restricts the further advancement of ultra-dense networks.How to efficiently reduce the interference among cells is the main problem of interference management algorithm.This dissertation analyzes the interference management algorithm of the existing ultra-dense network and focuses on the co-tier interference and cross-tier interference problem in the ultra-dense network.The main work of this dissertation is as follows:(1)In view of the co-tier interference among Pico Cells in ultra-dense networks and the shortcomings of existing clustering algorithms that can not effectively reduce interference within clusters,an improved clustering algorithm based on interference graph is proposed.Firstly,the interference graph is constructed by defining the weights of clusters based on the interference relation among Pico Cells.Secondly,according to the clustering algorithm,Pico Cells randomly distribut in the ultra-dense network are grouped into clusters.Considering that the size of the cluster affects the allocation of channel resources,the threshold value is set so that the Pico Cells can be evenly divided into each cluster.Finally,the interference management of Pico Cells is carried out dynamically by combining channel allocation and power allocation algorithm.Simulation results show that the proposed algorithm can reduce the interference among Pico Cells and improve the throughput of the system.(2)Aiming at the problem of cross-tier interference and co-tier interference among different types of cells in ultra-dense networks,the interference management algorithm of Pico Cells and Femto Cells cooperating and the interference management algorithm of Femto Cells based on interference graph coloring are proposed.First of all,aiming at the cross-tier interference among Pico Cells and Femto Cells,Pico Cells selects the cooperative base station according to the channel status information from Femto Cells to Pico Cells service users and the idle state information of Femto Cells.The selected cooperative base station sends the data information to the user of the Pico Cells service.Secondly,according to the interference strength of Femto Cells to users,it is divided into strong interference and weak interference,and Femto Cells with strong interference are divided into a cluster.Finally,the color of the interference graph is assigned to the vertices in each cluster according to the order of priority of the cost function.Meanwhile,the case that new Femto Cells request access to neighboring clusters is considered in this dissertation,and the neighboring clusters respond to the access request and decide whether to accept new Femto Cells according to the total weight value in the cluster.Simulation results show that the proposed algorithm not only ensures user fairness,but also improves the system throughput and system spectrum efficiency.