Research On Key Technology Of Vertical Handoff In Heterogeneous Wireless Networks Environment

In the past decade,wireless network obtained the rapid development,all kinds of new type of wireless access technology emerge in endlessly,which has brought unprecedented convenience to human life.But services offered by network with different access technology is not the same,not a single network can be completely replaced by other networks,so the fusion between heterogeneous wireless networks become inevitable.As the key part of integration and collaboration in heterogeneous wireless network,the research of vertical handoff has great significance on the optimization of wireless network resources and providing seamless service for users.The integration of heterogeneous wireless network leds to user's continuous improvement of quality of service requirements,whereas the existing vertical handoff algorithm used most of decision-making algorithm which based on strategy or multiple attribute,the parameters which these algorithms involved can be described through accurate model,but with the increasingly complex of network environment,previously established algorithm model may not apply to the change of the network,the realize of vertical handoff can only be achieved through a lot of manual intervention.This article uses the combined algorithm by bionics and neural network to overcome the above problems,in addition,our algorithm increase the management of the network load,can be regarded as a good way for a vertical handoff.In this paper,in order to bring a variety of wireless access technology integrated into a unified network environment,to achieve the target that effective utilization of entire wireless network resources and provide users with more seamless roaming service,we make the following work : 1)Establish common heterogeneous wireless network simulation platform,mainly used to simulate the real network environment,and provide experimental data and verify the performance of different vertical handoff algorithm;2)Establish preliminary decision models of vertical handoff,mainly used to reduce the system complexity and reduce the probability of occurrence of ping-pong effect;3)On the basis of the preliminary decision,we use ant colony algorithm and neural network to establish combined algorithm--ACO-BPNN(Ant Colony Optimization-Back Propagation Neutral Network)algorithm.The algorithm can avoid BPNN algorithm's defects that easy to be trapped in local minimum,at the same time can give consideration to the network load management,to realizes the intelligent seamlesshandoff.The simulation platform that established in this paper satisfied the demand of laboratory condition for heterogeneous wireless network.The simulation experiment results which obtained from the simulation platform shows that vertical handoff algorithm proposed in this paper can ensure low unnecessary handoff times,at the same time can meet the needs of the users for network resources.The proposed algorithm can also realize the effective utilization of network,so avoid network appear the phenomenon of heavy load to realize further intelligent vertical handoff.