Research On Clustering Based Interference Management In Ultra Dense Network

Ultra-dense networking is one of the key technologies of 5G.By deploying a large number of small base stations to form an Ultra dense network(UDN),seamless signal coverage is achieved,and the frequency resource reuse efficiency is improved.However,the intensive deployment of small base stations also brings some challenges,including interference management,mobility management,resource allocation and other issues.Because the interference environment under the ultra-dense networking is more complex than the traditional cellular network,the interference between the small and medium base stations in the UDN becomes more serious and turns into the main factor that restricts the network capacity.Based on the research status of UDN,this thesis studies the issue of UDN downlink interference management.The main research work is as follows:Firstly,this thesis introduces the basic concept,network architecture and channel model of UDN,and analyzes the defects of commonly used interference management technology in dense network environment.Then,the small base station clustering algorithms in UDN and the commonly used interference management technologies are investigated in detail.Secondly,aiming at the problem of co-tier interference between small base stations of UDN,this thesis proposes a joint clustering and channel allocation interference management scheme.This scheme introduces a weight value that measures the degree of interference between small base stations.By setting a weight threshold,it can determine non-negligible interference between small base stations and model the interference relationship between small base stations as a weighted undirected interference graph.Then,the optimization problem of maximizing system throughput is established,and the optimization problem is converted into two sub-problems of small base station clustering and channel resource allocation.The clustering problem is based on the weighted interference graph,and the heuristic clustering algorithm is used to cluster the vertices in the graph by maximizing the inter-cluster interference.The channel resource allocation is to allocate orthogonal subchannels for different clusters,that is,subchannels that maximize the cluster's throughput are assigned to each cluster by the subchannel assignment algorithm.Simulation results show that the interference management scheme in this thesis can improve the performance of the system.Finally,aiming at the problem of weak interference accumulation in UDN,this thesis proposes a hypergraph-based interference management scheme.In UDN,hyper-graph is used to model weak and strong interferences between small base stations,and optimization problems that maximize system throughput is established.According to the interference relationship in the hypergraph,the small base station is clustered using a greedy coloring algorithm,and sub-channels are allocated to small base stations in the cluster according to the user's QoS requirements.Simulation results show that the UDN interference management scheme based on hypergraph proposed in this thesis can achieve better system throughput on the basis of reducing interference.