Admission Control For Heterogeneous Wireless Networks Based On Dynamic Network Attributes And Non-cooperative Game

With the rapid development of wireless networks and mobile communicationtechnologies in recent years, lots of different wireless communication systems have beenappeared to provide services for users. Because these networks have differences on aspect ofcoverage, transfer rate, mobolity support and so on, the coexistence of these networks formheterogeneous wireless networks environment. The integration of heterogeneous networkswill be one of development trends of the next generation network. At present, the researchesabout heterogeneous networks integration are mainly focused on call admission control,vertical handoff, resource management, network selection, etc. Call admission control is oneof key technologies in the integration of heterogeneous networks in order to manage resourceand ensure QoS. The main contents in this article are as follows:Firstly, under the scenario of heterogeneous wireless networks consisted of WLAN andWiMAX, call admission control mainly includes network selection and network resourceallocation. At first, a network selection algorithm based on dynamic network attributes isproposed. In the algorithm, the network attributes can be changed dynamically according tonetwork load and the access network is selected through the method of Simple AdditiveWeighting. Then, based on considering the priority of different services and different calltypes, non-cooperation game model is used to solve bandwidth allocation between user andnetwork. Through solving the equilibrium of game, the final decision that the access networkadmits the user or not will be obtained.Secondly, in order to further reduce call request blocking probability and improvenetwork revenue, some improvements about admission control algorithm for heterogeneouswireless networks based on dynamic network attributes and non-cooperative game areproposed. The improved algorithm according network’s overall merit will select everynetwork in proper order to enter the phase of resource allocation, the call request is blockeduntil all networks cann’t meet the requirements of call request.Thirdly, we analyse admission control algorithm’s performance index in linking level intheory through establishing Markov Model. The performance index is call request blockingprobability. In simulation experiments, the blocking probability of call request with differentservices and different call types is analysed. At the same time, relevant algorithms in thisarticle are compared from several aspects, such as call request blocking probability, thenumber of accepted users, network revenue and bandwidth utilization.