Location Service in Mobile Computing Environments Ulf Leonhardt, Jeff Magee, and Paul Dias Dept. of Computing, Imperial College, 180 Queen’s Gate, London SW7 2BZ, UK {ul, jnm, pbd}@doc.ic.ac.uk With the advent of mobile computing devices and cheap location sensing systems, location information has become an important resource both for mobile and ‘desktop users’. In this paper, we describe some key concepts a scaleable ubiquitous location service should be based on. Firstly, we show how such a service can accommodate multiple location sensing systems. Secondly, we discuss hierarchy-based access control policies as a flexible and powerful mechanisms to protect users’ privacy. Thirdly, we address some issues concerning the visualisation of location information. 1 Introduction With users and computing equipment becoming increasingly mobile, location information is vital for many purposes, such as communication channel setup, navigation, and systems management. We envisage that environments that have a high proportion of mobile computing and communication devices will employ a general ubiquitous location service to cater for a wide spectrum of applications. The very basic services provided by such a system are: • to find the location of a specified located-object. • to find all located-objects at a given location. Both types of queries should be answered with a well-defined level of confidence. The location of objects can be measured in different ways for different types of objects. Some methods will only be available in certain geographic areas. Hence, the input of various sources of location must combined in order to achieve ubiquitous service coverage. Moreover, a higher (spatial and temporal) positioning accuracy can be achieved by exploiting overlaps among sightings from different sources. Statistical and other models (cf. Liu’s work [2]) can be also used to interpolate and extrapolate locations. In this paper, we argue in favour of combining all the location information that is available from the various primary sensing systems, along with mobility patterns and other heuristics, under the roof of an abstract location service that provides transparency from the hardware and software that actually gathers the sightings. Such a service would support context-aware applications (as described by Schilit in [18]), as well as location- aware services (such as Advanced Traveller Information Systems [19]). We describe an abstract framework that will be capable of accommodating arbitrary location sensing systems. The automated gathering of location information potentially infringes people’s privacy if too much data is disclosed to the wrong party. Therefore, our work investigates mechanisms to exactly specify and supervise the level of access to location data that is wanted. 1 Finally, we wish to support non-local roaming, and non-local location lookup. Essentially, we aim at the notion of global coverage for the location service. We investigate an architecture of federated location servers to build a global distributed location service. In section 2 we will describe our conceptual framework for the location service, followed by a discussion of the issues of security and server federation in sections 3 and 4, respectively. Finally, we address the visualisation of location information in section 5. 2 Conceptual Framework Definition: A located-object is an object whose location is of interest to the location service. 1 However, we cannot solve the underlying problems of society with our technology.