Interference Management Algorithm Based On Association Control In Heterogeneous Networks

In order to meet the growing demands of users for network,the deployment of some low-power small cell base station(BS)under the coverage of macro cell BS improves the overall access capability of network.While the deployment of small cell BS has brought unprecedented advantages to heterogeneous network,the spectrum reuse between the macro cell BS and small cell BS also greatly improves the spectrum efficiency.However,spectrum reuse could result in serious cross-tier interference,reducing the system gain.The association control could be used to reduce cross-tier interference and improve system throughput by changing the users' access mode.The traditional association control only allows the user to access one base station.In this thesis,the control condition is not controlled so strictly that the user is able to access two base stations at the same time.The problem of cross-tier interference could be addressed by accessing different number of base stations for users in the research system.When taking the fairness for users into consideration,this thesis chooses mathematical modeling by means of resource allocation and association control to optimize the overall throughput of the network.The original problem could be decomposed into two subproblems of resource allocation and association control.Firstly,corresponding resource allocation variables are set for different users in the network.By scaling the resource allocation variables and using Lagrange multiplier method and KKT conditions to solve the linear optimization,the optimal resource allocation scheme is chosen.Secondly,on the basis of optimal resource allocation,the greedy algorithm is then used to change the user's access network dynamically to achieve interference transfer.Finally,the algorithm of this thesis is compared with the current interference management algorithm to make a simulation comparison.The simulation result shows that the user access scheme proposed in this paper enables a quite marked improvement for the system throughput.