Wireless connectivity to communications and information has advanced the world towards ubiquitous computing. In the space of less than thirty years, cell phones have become ubiquitous and wireless data access has become common. However, this access has brought with it a variety of technical problems. Radio physics and power constraints, the need to reuse spectrum, economic constraints on facility placement, and service balkanization due to competitive and political factors force us to implement wireless systems as cells of limited range. Furthermore, cells may use very different wireless technologies or provide fundamentally different services, such as VoIP (Voice over IP), streaming, or direct short-range communications for telematics. We then need handoff mechanisms, often in multiple protocol layers, to allow a mobile terminal to move from cell to cell and maintain service continuity.
Mobility can be described as movement of a terminal, resulting in the release of the terminal's binding to the current cell (point of attachment to the network) and the establishment of bindings to the new cell being entered, while preserving the existing sessions associated with higher-level services. The cellular telephony community has long implemented service- and technology-specific mobility protocols that hand off voice sessions as the user moves from cell to cell. Because voice service quality is highly sensitive to service interruptions, cell-to-cell handoffs in a cellular environment have been highly optimized and are not noticed by the public. Tripathi et al. (1998) summarized some of the handoff technologies associated with cellular mobility. Pollini (1996) discussed some of the trends in handover design in cellular networks that may affect the optimization of handover performance.
For IP traffic, the IETF has defined mobility protocols for both IPv4 (Perkins, 2002b) and IPv6 (Johnson et al., 2004). However, IP traffic is dramatically more diverse than cellular voice in the range of link layer technologies used to support IP traffic, the number of economic units supplying IP services, and the authentication protocols and services running above IP. This diversity has meant that the IETF could not easily design access-specific handoff optimization techniques such as soft handoff (Chen and Mary, 2003; Wong and Lim, 1997), often seen in cellular voice, into the mobility standards. As a result, unoptimized Mobile IP handoffs can take a few seconds to perform, and degrade the quality of service in the process.
IP's transformation from a service supporting email and file transfer to the base layer for network convergence means that the constraints on handover performance are becoming much more stringent. Handovers cannot interrupt real-time services. The mechanisms and design principles needed for building optimized handovers in the context of mobile Internet services are poorly understood and need better analysis. To the best of our knowledge, none of the existing work has attempted to model the systems aspects of a mobility event nor was intended to systematically analyze the elementary operations involved in a mobility event. This body of work is also lacking in the ability to predict the performance of a mobility event. Some of the existing work has focused on optimizing only parts of a mobility event in an ad hoc manner, specific to a mobility protocol, without providing a comprehensive approach to solving the optimization problem in all layers or functional modules. We provide an overview of this related work for each of these mobility functions along with a detailed description of the proposed techniques in Chapter 6.
This book is intended to contribute to a general theory of optimized handover, especially with respect to the mobility of Internet-based applications. The contributions fall into four categories:
- Identification of fundamental properties that are rebound during a mobility event. Analysis of these properties provides a systematic framework for describing mobility management and the operations that are intrinsic to handover.
- A model of the handover process that allows one to predict performance both for an unoptimized handover and for specific optimization methodologies under conditions of resource constraints. This model also allows one to study behavioral properties of the handoff system such as data dependency and deadlocks.
- A series of optimization methodologies, experimental evaluations of them, and optimization techniques that can be applied to the link, network, and application layers and preserve the user experience by optimizing a handover.
- Application of the model to represent optimizations, and comparison of the results with experimental data.
1.1 Types of Mobility
There are several types of mobility, such as terminal mobility, personal mobility, session mobility, and service mobility. Schulzrinne and Wedlund (2000a) introduced several different types of mobility to support multimedia traffic for an IP-based network. We briefly review each type of mobility.
1.1.1 Terminal Mobility
Terminal mobility allows a device to move between networks while continuing to be reachable for incoming requests and maintaining existing sessions during movement. It allows an established call or session to continue when an MS (mobile station) moves from one cell to another without interruption in the call or session.
Terminal mobility may also arise from a change in the network condition, whereby a mobile may switch between two neighboring networks even without any movement. We describe here the types of terminal mobility that arise from different types of handoffs. Handoff, often also known as handover, is a process that results when a mobile disconnects from one point of attachment in a network and reconnects to another point of attachment in the same or a different network.
The handoff process can be either hard or soft. With hard handoff, the link to the prior base station is terminated before or as the user is transferred to the new cell's base station. Thus, the mobile is linked to no more than one base station at a given time. Initiation of the handoff may begin when the signal strength at the mobile received from the target base station is greater than that from the current base station. As the mobile moves into a new cell, its signal is abruptly handed over from its current cell (or base station) to the new one. In the old analog systems, hard handover could be heard as a click or a very short beep. In digital cellular systems, this is not noticed. However, in an IP-based handoff scenario, hard handover contributes 4–15 s of delay (Dutta et al., 2005c). With soft handoff (Wong and Lim, 1997), the MS continues to receive and accept radio signals from the base stations that are part of the previous as well as the new cell for a limited period of time. The MS signal is also received at multiple base stations. In order to ensure the layer 2 independence requirement of mobility management schemes, a maximum acceptable handoff time (MAHT) is required, which will vary based on the access type.
In an end-to-end wireless IP environment, four logical levels of handoff procedures can be defined:
- Layer 2 handoff. This allows an MS to move from one layer 2 point of attachment to another layer 2 point of attachment that belongs to the same subnetwork. Each layer 2 point of attachment may be equippedwith same or a different type of radio access technology. One subnetwork may consist of multiple layer 2 radio access networks. The IP address of the mobile host remains the same during this handoff.
- Subnet handoff. This allows an MS to move from a radio access network within a subnet to an adjacent radio access network within another subnet that belongs to the same administrative domain. The IP address of the mobile may or may not change.
- Domain handoff. This allows an MS to move from one subnet within an administrative domain to another in a different administrative domain. Domain handoff can...
