Location Prediction Of Mobile Group Based On Apriori Algorithm In The Ubiqitous Network

Recently,with the continuous development of the mobile communication technology and mobile terminal devices,the user can not be satisfied by using single terminal and single network.And the multi-terminal and multi-network collaboration will be the direction of the future communication.Multiple networks form the ubiquitous network.In the ubiquitous network,quality of service (Qos) is difficultly guaranteed for the particularity of the ubiquitous network network. The location prediction can be applied in mobility management,call admission control,smooth handoffs and resource reservation for improved quality of service.If the system knows, prior to the mobile subscriber movement,the exact trajectory it will follow,the Qos can be improved. Thus,location prediction is the key issue to provide quality of service to mobile subscriber.Moreover,recognizing the multi-terminal and multi-network collaboration,a location prediction proposal of mobile group based on Apriori algorithm in the ubiqitous network is proposed.There are three key points for the proposed location prediction proposal.First, mobile behavior of users is decided by people themselves.However,in the daily life,mobile users often repeat regular routes in certain periods.Apriori algorithm is used to mine mobile users’mobile model.This information is used to predict mobile users’future locations. Simulation results show that the average prediction accuracy ratio of the location prediction scheme based on Apriori algorithm proposed in this paper is about21.73%higher than that of the based fist-order Markov model location prediction scheme.Second,the collaborative mobile group constitutes a virtual terminal.the terminal group,which constitutes a virtual terminal,and the main control terminal are obtained,by using technique for order preference by similarity to an ideal solution.The location prediction is carried out by using the location prediction scheme based on Apriori algorithm proposed in this paper for the main control terminal,the other terminals in the group share location prediction information with the one,whose location prediction information has been known,by the way of personal area network.so it not only can realize the location prediction of mobile groups but also can reduce the time overhead and the space cost of the mobile group.Third,a location prediction mechanism based on logistic cells(LPMboLC) is used to shield the differenc on network cells.This mechanism makes that almost every existing mobility prediction algorithm can be implemented on this LPMboLC mechanism in heterogeneous wireless networks.