Improving Positioning Accuracy in WAW Location-Based Services José E. Sanguino 1 , Filipe Tocha 2 , António Rodrigues 3 IT – Instituto de Telecomunicações Instituto Superior Técnico, Technical University of Lisbon, Portugal 1 sanguino@lx.it.pt 2 ftocha@gmail.com 3 antonio.rodrigues@lx.it.pt Abstract—Location-based services rely on the accuracy of the user’s position estimates. In WAW (“W here A re W e?”) location- based services, users want to be aware, not only, of their own position but also the position of the other elements on their group. If this information is available in real-time, this kind of service can be used to support coordinated positioning operations. This paper presents an algorithm for the estimation of a user’s position, based on measurements from a group of mobile Global Positioning System (GPS) receivers. Receivers are assumed to have some common satellites in view, but measurements from all the satellites are used in the estimation process. Instead of estimating the user’s position on an autonomous way, based only on the measurements of his own receiver, the proposed approach takes advantage from all the GPS measurements available in this kind of location-based service. None of the GPS receivers, used in the estimation process, is considered to be fixed or known. The approach leads to an improvement in the user’s position accuracy and can particularly benefit those with a more restricted view of the satellite constellation. As such, the methodology, here presented, can be applied with advantage in the deployment of WAW location-based services in urban environments, more prone to the canyon effect. The paper describes in detail the positioning model and estimator used to achieve the accuracy improvements. Simulation results are presented based on real GPS satellite ephemeris, collected at the university campus. I. INTRODUCTION Location-based services explore the synergies between radionavigation and mobile communication systems. In coordinated positioning operations each team member requires, in real-time, not only, his position but also the position of all the members on his team. Examples of such kind of operations can go from manoeuvring troops in a battlefield, to search-and-rescue teams in catastrophe scenarios, or from a group of vehicles on a convoy journey across the country, to a group of friends spending a day at a ski resort or golf field, [1]. Most of this kind of coordinated positioning operations can be supported on the services provided by a WAW (“W here A re W e?”) location-based service, where the position of each user is made available to the rest of his group. These services can be developed by combining the positioning capabilities of satellite navigation systems, such as GPS, with the communication capabilities of mobile communication systems. It is from this merging of navigation and communications that arises the added value of WAW location-based services. The most basic form of a GPS-based WAW service is supported on the exchange of user’s positioning coordinates, such as latitude, longitude and altitude, estimated locally by each GPS receiver. Pseudoranges are measured by each receiver, to a set of visible satellites, and used in this estimation process. In an autonomous way, each receiver uses only its measurements to estimate its own position. The positioning accuracy depends, among other factors, on the number of visible satellites and their relative position, [2]. Two users with a different view of the satellite constellation are expected to experience different positioning accuracies. Mobile users in urban environments should experience different accuracies as their view of the sky changes as they move. The possibility of using the measurements from several GPS receivers, which exists in the WAW location-based services, motivated the development of the positioning algorithm proposed in this paper. By using the measurements from several receivers, to common satellites, it is possible to estimate the user’s relative position with high accuracy, [3]. However, the goal is to increase the absolute positioning accuracy. Thus, in the proposed approach, these measurements, that allow a high relative positioning accuracy, are merged into the absolute positioning algorithm, leading to an improvement in the absolute positioning accuracy. All the receivers involved in this estimation process benefit from the accuracy gain. This gain can particularly benefit those receivers with a reduced view of the satellite constellation. As long as common satellites exist, a receiver with a reduced view of the constellation can still be positioned with high accuracy, relatively to the other receivers. By this way, the estimation of its absolute position benefits from the better view of the satellite constellation experienced by the other receivers. In the estimation process none of the receivers is considered fixed or with a known position. The rest of the paper is organized as follows. In section II, the system architecture for WAW location-based services is briefly described. In section III, a detailed description of the proposed positioning model and estimator is presented. Simulation results, based on real GPS satellite ephemeris, are presented in section IV and illustrate the performance of the proposed algorithm. Section V closes the paper, with conclusions and remarks. 978-1-4244-2489-4/08/$20.00  2008 IEEE IEEE ISWCS 2008 123