Research On Vertical Handover Technologies Of Wireless Heterogeneous Networks

With the rapid development of wireless communication technologies, wireless heterogeneous networks become popular quickly. Vertical handover technology as the key technology in research on wireless heterogeneous networks has got a lot of attention all the time. In recent years, the rapid development of mobile Internet services puts forward new requirements to traditional vertical handover technologies. Vertical handover technology should not only make dynamic network choices based on the real-time status of terminals and networks, but also protect users'location privacy in mobile networks.This dissertation intends to solve the dynamic network selection problem and the location privacy protection problem in vertical handover. Based on previous studies, a vertical handover algorithm based on Dynamic Fuzzy Analytic Hierarchy Process (DF-AHP) and a location privacy protection technology in vertical handover are proposed in this dissertation. Meanwhile, they are implemented and tested in the actual environment.Firstly, a vertical handover algorithm in heterogeneous networks based on DF-AHP is proposed in this dissertation. This algorithm adds dynamic update functions to traditional AHP so that elements in the judgment matrix can be dynamically changed according to the dynamic elements. Because of this, this algorithm can meet the needs of dynamic network selection in heterogeneous network environments. Meanwhile, the triangular fuzzy number is used in this algorithm to solve the problems that the consistency analysis in traditional AHP is too cumbersome and the resolution of weights may be too small.Secondly, a location privacy protection technology in vertical handover is proposed. This technology consists of three parts:an assisted independent system model, the mixing zone generating module and the fake users generating module. The assisted independent system model can collect the handover record and it is the basis of this technology. The mixing region generating module can form clusters by using limiting density-based spatial clustering of applications with noise algorithm (L-DBSCAN). L-DBSCAN limits the size of every cluster by setting the upper limit of the number of the targets in a cluster so that it can significantly enhance the number of clusters and the size of individual clusters will not be too big on the premise of clustering. The fake users generating module generates fake users whose location coordinates are subject to the two-dimensional Gaussian distribution in every mix zone. Moreover, this technology allows users to define the number of fake users by themselves.In the end, the two methods mentioned above are proposed are implemented and tested in the actual environment. The results show that the vertical handover algorithm based on DF-AHP can make real-time dynamic choices according to external conditions and the fake users produced by L-DBSCAN can protect users'location privacy effectively. And compared with DBSCAN, L-DBSCAN has better performance of location privacy protection.