Research On Network-based Mobility Management Mechanisms

Mobile IP(MIP)enables mobile node to maintain its connectivity to the Internet during handover.However,mobile IP protocols are mobile node centric,which the handover related decision making is mostly done in the mobile node only.Other IP mobility management protocols such as Host Identity Protocol(HIP),Mobility and Multihoming extension to Internet Key Exchange(MOBIKE),Stream Control Transmission Protocol(SCTP)are all mobile node centric,in which there are no network side's entities supporting handover decision,mostly the decision made by mobile node.Existing mobile IP technology are terminal based mobility scheme,which is difficult to fulfill the diverse mobility management requirements of future heterogeneous mobile network.A solution for an Internet mobility architecture that could be widely deployed consists of the combined use of a host-based global mobility protocol to operate between multiple access network providers and a network-based localized mobility management protocol to optimize the operation of networking within a single access provider's network.Therefore,standards development organizations such as IETF,IEEE,WiMAX,3GPP,3GPP2 and ITU-T have began the requirements analysis and protocol research work of network based mobility management.Their efforts identify that it is very necessary to study the network based mobility management technology.In this dissertation,the network based mobility management related key problems are studied.First,the related research advance and tendency about host based and network based mobility management are summarized,and our research considerations and work are presented.And then five problems which are a protocol for network-based localized mobility management solution,a fast handover scheme in Proxy Mobile IPv6,an authentication method in Proxy Mobile IPv6,a Proxy Mobile IPv6 based Global Mobility Management architecture,and a split mechanism based Internet mobility management are studied respectively.Meanwhile,relevant protocols and analysis models are presented,and the studied results of analysis and simulation are given.The main contributions of this thesis are shown as following:(1)We introduce a protocol for network-based localized mobility management solution, and develop an analytic model for the performance analysis based on one-dimensional random walk of mobile node.Based on the analytic models,the location update cost and the packet delivery cost are formulated.Then,the impacts of average cell residence time and the number of mobile nodes in a cell on the total cost are analyzed,respectively.In addition,the variation of the total cost is studied as the session-to-mobility ratio is changed and the optimal local mobility domain size to minimize the total cost is also investigated.The analysis results indicate that the session-to-mobility ratio and the local mobility domain size are critical performance factors in order to minimize the total cost.(2)We propose a fast handover scheme in Proxy Mobile IPv6 called FPMIPv6,which combines Fast Mobile IPv6 operation and IEEE 802.21 link layer triggers with Proxy Mobile IPv6 protocol,and develops an analytical model for the handover latency performance analysis.Then,the impacts of various parameters on the handover latency are analyzed,respectively.After the comparison with the existing Proxy Mobile IPv6 scheme,it can be seen that the proposed fast handover scheme can achieve low handover latency.(3)We present an authentication protocol in Proxy Mobile IPv6.Proxy Mobile IPv6 provides a means of network-based mobility management for mobile nodes without involving the mobile nodes themselves in the mobility signaling.To make Proxy Mobile IPv6 protocol feasible in mobile Internet,a new defined authentication protocol should be accompanied.In addition to providing access authentication,the proposed authentication protocol prevents threats such as replay attack and key exposure.Also,we develop analytic models for the authentication latency and cost analysis.Then,the impacts of mobility and traffic parameters on the authentication cost and latency are analyzed,respectively.The results show that the proposed authentication protocol for Proxy Mobile IPv6 is secure and effective.(4)We specify a Proxy Mobile IPv6 based Global Mobility Management architecture and protocol procedure.In the proposed architecture,the mobility management is performed by network entity other than mobile node itself.The other benefit is the elimination of the wireless link data delivery tunnel overhead between the mobile node and the access router.To compare with well known hierarchical mobile IPv6 mobility management protocol,the location update,the packet delivery and total cost functions generated by a mobile node during its average domain residence time in each protocol are formulated based on fluid flow mobility model.Then,the impacts of various system parameters on the cost functions are analyzed, respectively.The analysis results indicate that the proposed proxy mobile IPv6 based global mobility management protocol can guarantees lower total costs.(5)We present a new Internet architecture based split mechanism and its mobility management protocol.It cleanly separates host location from its identity and explicitly supports mobility by separating the transport layer function from the network layer function in addresses.It cleanly separates the Internet access networks from the core network.It makes IPv6 routing both in access networks and the core network more scalable.Specially,this paper gives the mobility management protocol of the proposed Internet architecture.It supports host based and network based mobility management.Also,we develop analytic models for the mobility management costs and handover latency.Then,the impacts of mobility and system parameters on mobility management costs and handover latency are analyzed, respectively.The results reveal that the handover latency of the proposed mobility management protocol is bigger due to location management messages across more network process nodes and the network based mobility management is better than the host based mobility management in handover latency.