Resource Allocation In Small Cell Network Based On Dynamic Clustering And Game Theory In The Context Of 5G

With the development of mobile internet technology,the popularity of smart mobile terminals as well as the rise of a variety of emerging businesses,the user's demand for mobile traffic is getting higher and higher,and the scarcity of available spectrum,which is a huge challenge for the current communication network.Small cells deploy flexibly and can provide users with close and high-quality services,which can provide a good idea to improve the spectrum utilization and system capacity,and it is expected to become a key technology of the new generation of communication system.However,it will bring many challenges as the deployment of small cells,including the planning and management of small cells,and the interference problem when small cells and macro cell use the same frequency,so it is a key technology to coordinate,and that is the focus of current research too.Resource allocation is an important coordination technology,including spectrum allocation,power control and access allocation,which can effectively reduce the interference and improve the performance of the system.This paper mainly study the spectrum allocation of small cell network,and introduce clustering and game theory,to find the optimal spectrum allocation of small cell network,improve the spectrum efficiency,reduce interference and improve the throughput of the system.The main work of this paper includes:(1)In this paper,a spectrum allocation scheme based on clustering and non cooperative game is proposed in the two layers network with small cells and macro cell.First of all,in the scenario with intensive small cells,aiming at the problem of large numbers,strong autonomy and difficulty in centralized management,we propose a hierarchical clustering algorithm based on K-MEANS,which clusters the small cells based on the distance and traffic.Then,the frequency spectrum is divided into the central frequency spectrum and the edge spectrum by the method of FFR inner each cluster to reduce interference between cluster members.Finally,a non cooperative game whose players are clusters is constructed and a spectrum allocation optimization algorithm based on non cooperative game and correlation equilibrium is proposed,which can solve the correlated equilibrium of the game by regret matching mechanism.(2)A spectrum sharing algorithm based on coalition game is proposed,which each small cell is a player of the game and forms coalition with other small cells to reduce interference and increase utility.Inner each coalition,small cells share the conflict spectrum by the method of TDMA and obtain the corresponding utility according to the contribution of the shared spectrum.Under the framework of coalition game,we can get the stable coalition structure by setting the small cells switch operation and the superior relationship of coalitions,so as to optimize the spectrum allocation of small cells network and improve the performance of the system.