Traffic Based Network Access Strategies In Heterogeneous Wireless Networks

The rapid development of the wireless communication technologies has totally changed the world of communications in the past several decades.The heterogeneous wireless networks which integrate multiple wireless access technologies that are different in terms of their general characteristics is the inevitable trend to satisfy the growing traffic demands.How to help consumers with multiple kinds of terminals select one or more access networks and acquire proper radio resource with the consideration of the network information and users' preference is one of the critical issues in the heterogeneous wireless networks.The thesis firstly researches the single network selection strategies for the multi-mode terminals and propose a MADM and queuing theory based network selection algorithm.Then the multi-radio access and radio resource allocation methods for multi-homed terminals are studied.The thesis puts forward a TOPSIS based multi-radio access network selection method and multi-radio access based bandwidth allocation strategy for video communication.The mainly researches of this thesis are listed as follows:(1)MADM and queuing theory are applied to network selection decision and the thesis proposes a MADM and M/M/1/N based network selection algorithm.In this algorithm,the candidate networks are determined according to the satisfaction degree between actual network attributes and correspondly expected and the network coverage ability which lead by mobility.Moreover,the weights of each attributes in satisfaction degree function are determined by the modified AHP.Then develop the related matrix between networks and users and adjust the elements according to M/M/1/N queuing theory for the minimum of the average queuing delay.Simulation results show that the proposed algorithm in this paper can reduce the quequing delay of the system and promote the user handoff performance significantly.(2)For the multi-homed terninals which can simultaneously access to multiple networks,this thesis proposes a MADM based multi-radio access network selection strategy.The multi-homed terminals collect network condition information periodically and decide the candidate networks on the basis of RSS and threshold.The subsets of candidate networks are devided in terms of radio access technologies they adopted.Then TOPSIS that combines network load condition,cost,energy consumption and throughput are applied to evaluate each network in the subsets.The simulation results show that this method can provide higher throughput for users and improve the load balancing performance effectively.(3)A multi-radio access based bandwidth allocation strategy for video communication is proposed in this thesis.In this algorithm,the multi-homed terminals decide the available networks in accordance with RSS and choose the best networks from each technologies as the active set elements on the base of SAW.Then the video satisfaction model,cost model and energy efficiency model are build and linear weighted combined.The joint utility function are restricted to the video communication transmission rate and networks total bandwidth.Finally,the optimized bandwidth allocation are acquired according to Lagrange multiplier and KKT conditions.Simulation results show that the proposed algorithm promotes the network load performances and guarantees that users obtain the best joint utility and satisfied video communication experience.