Sajal K. Das, Amiya Bhattacharya, Abhishek Roy, and Archan Misra
Models of twenty-first century ubiquitous computing scenarios  depend not just on the development of capability-rich mobile devices (such as Web phones or wearable computers), but also on the development of automated machine-to-machine computing technologies, whereby devices interact with their peers and the networking infrastructure, often without explicit operator control. To emphasize the fact that devices must be imbued with an inherent consciousness about their current location and surrounding environment, this computing paradigm also is called sentient  (or context-aware) computing. Context awareness is one of the key characteristics of applications under this intelligent computing model. If devices can exploit emerging technologies to infer the current state of user activity (e.g., whether the user is walking or driving, at the office, at home, or in a public environment) and the characteristics of the user's environment (e.g., the nearest Spanish-speaking ATM), they can then intelligently manage both the information content and the means of information distribution.
Location awareness is the most important type of context, because the current (or future) location of users strongly influences their information needs. Applications in computing and communications utilize such location information in two distinct ways:
Location-aware computing: In this category, the information obtained by a mobile device or user varies with changes in the user's location. The most-common goal on the network side is to automatically retrieve the current or anticipated neighborhood of the mobile user (for appropriate resource provisioning); on the device side, the typical goal is to discover appropriate local resources. As an example of this category, we can consider the case where mobile users would be automatically provided with local navigation maps (e.g., floor plans in a museum that the user is currently visiting), which are automatically updated as the device changes its current position.
Location-independent computing: Here, the network endeavors to provide mobile users with a set of consistent applications and services that do not depend on the specific location of the users or on the access technology employed to connect to the backbone information infrastructure. Information about the user's location is required only to ensure the appropriate redirection of global resources to the device's current point of attachment; such applications are not usually interested in the user's absolute location but only in the point of attachment to the communications infrastructure. An example of this is cellular telephony, where mobility management protocols are used to provide a mobile user with ubiquitous and location-independent access.
While location-independent computing applications have a fairly mature history, location-aware computing is still at an early stage. Innovative prototypes of location-aware computing environments are still largely experimental and geared toward specific target environments. The location support systems of different prototypes, as a result, have been largely autonomous and have always remained at the disposal of the system designers. It is important, however, to realize that the full potential of location-aware computing can be harnessed only if we develop a globally consistent location management architecture that caters to the needs of both location-aware and location-independent applications, and allows the retrieval and manipulation of location information obtained by a wide variety of technologies. This is an interesting technological challenge, because location-aware and location-independent applications typically face significantly different scalability concerns. In general, location-aware applications do not generate significant scalability issues because they primarily involve local interactions; however, scalability is a critical concern for location-independent network services, which must support access to distributed content by a much larger user set.
In this chapter, we focus on identifying the various requirements that must be satisfied by such a universal location-management infrastructure. We also explain why we prefer that such location data be expressed in symbolic format, and then discuss the use of information-theoretic algorithms for effectively manipulating such location information. A symbolic representation of location data allows the management infrastructure to deal with an extremely heterogeneous set of networking technologies, with a wide variety of underlying physical layers and location sensor technologies. Indeed, the ability to accommodate device heterogeneity and technological diversity is vital to the success of a universal location-management scheme. As our survey of current trends will show, location information in various prototypes differs widely in their environment of applicability and the granularity of resolution. Moreover, we shall see how such symbolic information is more amenable to storage and manipulation across heterogeneous databases, and can be exploited to provide necessary functions such as location prediction, location fusion, and location privacy.
The rest of the chapter is organized as follows. Section 17.2 highlights some examples of location-aware applications and prototype systems. The various location-related functions that must be realized in a universal location-management framework designed for pervasive computing models are identified in Section 17.3. Based on this discussion, we explore also the relative merits of alternative schemes for global location representation. Section 17.4 discusses the novel concept of path update and outlines the LeZi-update algorithm that optimizes the signaling loads associated with location tracking. We present ongoing research that extends this algorithm to provide translation of location information across heterogeneous networks and multiple access technologies. Section 17.5 summarizes the chapter and discusses open problems.
Weiser, M., The computer for the 21st century, Sci. Am., 265 (3), 94–104, 1991.
Hopper, A., Sentient computing, The Royal Society Clifford Patterson Lecture, 1999, http://www-lce.eng.cam.ac.uk.