3608 IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. 8, NO. 7, JULY 2009 Proposal and Analysis of Adaptive Mobility Management in IP-Based Mobile Networks Rami Langar, Member, IEEE, Nizar Bouabdallah, Member, IEEE, Raouf Boutaba, Senior Member, IEEE, and Bruno Sericola, Senior Member, IEEE Abstract—Efficient mobility management is one of the major challenges for next-generation mobile systems. Indeed, a mobile node (MN) within an access network may cause excessive signaling traffic and service disruption due to frequent hand- offs. The two latter effects need to be minimized to support quality of service (QoS) requirements of emerging multimedia applications. In this paper, we propose a new adaptive micro- mobility management scheme designed to track efficiently the mobility of nodes so as to minimize both handoff latency and total signaling cost while ensuring the MN’s QoS requirements. We introduce the concept of residing area. Accordingly, the micro- mobility domain is divided into virtual residing areas where the MN limits its signaling exchanges within this local region instead of communicating with the relatively far away root of the domain at each handoff occurrence. A key distinguishing feature of our solution is its adaptive nature since the virtual residing areas are constructed according to the current network state and the QoS constraints. To evaluate the efficiency of our proposal, we compare our scheme with existing solutions using both analytical and simulation approaches for the 2-D random walk model as well as real mobility patterns. Numerical and simulation results show that our proposed scheme can significantly reduce registration updates and link usage costs and provide low handoff latency and packet loss rate under various scenarios. Index Terms—QoS provisioning, mobility management, hand- off algorithm, micro-mobility, MPLS, Mobile IP, adaptability, performance analysis. I. I NTRODUCTION F UTURE wireless networks are expected to provide IP- based coverage and efficient mobility support with end- to-end QoS guarantees. Two enabling factors are considered as crucial: (i) maintaining the network connectivity during node mobility and; (ii) provisioning the network resources required by the Mobile Node (MN) in all the visited subnetworks. Mobility management protocols are key for service continu- ity in mobile networks. Mobile IP [1], the Internet Engineering Task Force (IETF) standard, can serve as the basic mobility management in IP-based wireless networks. According to Mobile IP, a MN can change its point of attachment without changing its IP address. To do so, a MN is assigned with a permanent home address in its home network, and will Manuscript received February 25, 2008; revised September 30, 2008 and February 18, 2009; accepted March 30, 2009. The associate editor coordinating the review of this paper and approving it for publication was H. Zheng. R. Langar is with the Computer Science Laboratory of Paris 6 - LIP6, University of Pierre and Marie Curie, UPMC - Paris Universitas, 75016 Paris, France (e-mail: rami.langar@lip6.fr). R. Boutaba is with the Computer Science Department, University of Waterloo, ON, N2L3G1 Canada (e-mail: rboutaba@uwaterloo.ca). N. Bouabdallah and B. Sericola are with INRIA, 35042 Rennes, France (e-mail: {nizar.bouabdallah, bruno.sericola}@inria.fr). Digital Object Identifier 10.1109/TWC.2009.080166 borrow a temporary care-of address (CoA) in each foreign network. The CoA is the foreign agent (FA) IP address of the currently visited foreign network. In this case, the home agent (HA), residing in the MN’s home network, will maintain the mapping between the home address and the CoA. Clearly, such mechanism induces long handoff latency and large signaling load when handoffs occur frequently [2]. In this regard, many enhancements to Mobile IP for MNs with frequent handoffs have been proposed in the literature [3]–[11] to ensure service continuity. The second major factor identified as crucial for the evolu- tion process of future networks is efficient network resources provisioning. This issue has been largely studied in both wired and wireless environments. For instance, MultiProtocol Label Switching (MPLS [12]) addresses today’s network backbone requirements effectively by providing a standardized solution that improves packet forwarding performance and designs high performance QoS guaranteed paths. Specifically, MPLS forwards data using labels that are attached to each data packet instead of traditional IP destination lookup. These labels are distributed using a label distribution protocol, which maintains the coherence of label bindings across the network. MPLS with its traffic engineering (TE) attempts to provide a means to manage and enhance network traffic through rigorous analytical studies. As a matter of fact, there is an increasing trend towards the introduction of MPLS in wireless environments [13]–[18]. To meet the requirement of next generation mobile net- works, we propose in this paper a new adaptive micro-mobility management scheme called adaptive Master Residing Area (MRA) which alleviates the limitations of previous works in terms of flexibility and adaptability and in the same time benefits from MPLS resource provisioning capability. The key idea behind our proposal is to manage adaptively the node mobility according to its current state and the QoS constraints. To do so, we introduce a new concept called Residing Area (RA). The size of RAs is not the same all the time for a particular MN and is dynamically computed according to the two above conditions (i.e., user’s position and the QoS constraints). This allows reducing the signaling cost while respecting the specific QoS MN’s needs. To gauge the effectiveness of our proposed scheme, we develop a new analytical model based on Markov chains. We, explicitly, derive the expressions of the signaling cost function of registration updates, the link usage cost and the handoff performance metrics (i.e., handoff latency and packet loss rate) for a general two-dimensional (2-D) random walk mobility model. In addition, simulations are conducted using 1536-1276/09$25.00 c  2009 IEEE Authorized licensed use limited to: UR Rennes. Downloaded on January 5, 2010 at 11:57 from IEEE Xplore. Restrictions apply.