In this paper, we describe a queueing network model representing the Mobility Plane Architecture, which allows mobile users to register in the system and move around while stay connected. This paper presents a system overview, including node attachment, node updating and handover procedures. Two queueing network models and their evaluation on a simulation tool are presented, resulting in some important conclusions for our architecture and other micro-mobility proposals. Results show the importance of considering L2 and L3 association times because they can affect all procedures involving new registration and handover.
I. INTRODUCTIONThe Internet Engineering Task Force (IETF) published several standards for supporting mobility in IP networks. The standards are divided into two categories, those supporting macro-mobility and those supporting micro-mobility. Macromobility refers to mobility across several networks, while micro-mobility refers to mobility within a single and potentially large network.Micro-mobility protocols aim to improve localized mobility by reducing the handover overheads [1]. Existing solutions for IP mobility present strong drawbacks that make these solutions simply undeployable. The problems are related to the complexity of the proposed solutions that turn their implementation on small mobile devices such as cell phones and handhelds unfeasible. In fact, manufactures of such devices never considered supporting the present solutions.Several projects aim to design and prototype architectures for supporting mobility in IP networks. The Mobility Plane Architecture Project [2] supports micro-mobility in IP, MPLS and GMPLS networks. Support for micro-mobility is a key issue in mobile IP networks since it speeds up the handover process, minimizing communication disruptions when a mobile node changes its network point of attachment.The problem addressed by this architecture is how to support micro-mobility in standard IP or MPLS/GMPLS networks in a network-centric way, that is, the burden demanded by micro-mobility is placed on the network, not on the mobile nodes.The aim of this paper is to present presents an overview of MPA architecture and some related models and performance results. A prototype of this architecture has been built, from