Research On Call Admission Decision And Control Algorithm In Heterogenous Networks

With the development of networks and wireless technologies, the integration of variousheterogeneous networks has drawn more and more attention within these years, which canultimately realize the ubiquitous network society. Ubiquitous network makes anyone can easilyaccess and use a network at anytime from anywhere and from any appliance. To guarantee the QoS,the reasonable call admission control policies are needed in heterogeneous networks.This thesis investigates the call admission control algorithm in heterogeneous networks. Itinvestigates a call admission control policy based on Markov Decision Process (MDP) to facilitatethe evaluation of call admission control policy for different multiclass traffic in heterogeneouswireless networks.Firstly, the thesis discusses the system model of a heterogeneous wireless network architectureconsisting of LTE network and WLAN. The parameters of the model are determined. The thesisdefines multiple call classes (i.e., new calls, vertical handoff calls and horizontal calls) where callsmay have single or multiple variable channel requirements. Then, the thesis describes the fivecomponents of the MDP model.Secondly, the simulation of this call admission control policy is made. In order to determinethe optimal decision, the Value Iteration Algorithm is introduced, which can calculate the expectedcall admitted to the system. Furthermore, the thesis considers simulation result for three differentweighting factor set.At last, the thesis evaluates the performance of the proposed algorithm according to thesimulation result. It analyzed the performance of each class with different channel requirements.Simulation result shows that calls dropping rate completely depend on their own priority levelswhich differ in different networks. In addition, calls with lower channel requirement have higherpriority to admit into the system.