Efficient Proximity Detection for Location Based Services Georg Treu and Axel K¨ upper Mobile and Distributed Systems Group, Institute for Informatics Ludwig-Maximilian University, Munich, Germany e-mail:[georg.treu|axel.kuepper]@ifi.lmu.de Abstract - Proactive location-based services (LBSs) are characterized by the fact that (spatial-related) events are automatically triggered without explicit user invo- cation. An example is the notification of users when entering a certain zone. Since continuous tracking of a target is not feasible due to the limitations of the air- interface, efficient position update strategies have to be conceived. For relating the position of a mobile terminal with fixed geographic entities, such as points of inter- est, relatively simple strategies are sufficient. In this paper an efficient but simple update strategy for detecting proximity between mobile terminals is presented. 1 Introduction Proactive location-based services (LBSs) are characterized by the fact that they are automatically triggered if a pre-defined spatial-related event occurs, for example if a target enters a certain zone and the LBS user wants to be notified about that. In today’s networks, almost all LBSs are reactive. A prominent example is the buddy finder, where users can request the position of their buddies (assuming that each buddy agrees to that). If the user invokes this service, the positions of all targets are collected. Proactive LBSs are scarcely available today, one reason for that certainly being the lack of an efficient position management. The proactive complement to the buddy finder is a community service, which automatically notifies a member as soon as another member approaches, for example when the distance between them falls below a pre-defined threshold. To realize this service an efficient tracking strategy needs to be applied since naively tracking all community members continuously is too excessive with regard to the signaling overhead caused at the air- interface. The problem of detecting proximity between mobile terminals (MTs), parameterized with a given distance threshold, is a general one. Applications for which an efficient proximity detection service is fundamental range from the mentioned community service to mobile gaming, fleet management and others. In this paper we describe and evaluate a strategy for efficient proximity detection among MTs. For this purpose, we assume terminal-based positioning like GPS, i.e., the MT performs range measurements and derives its position from the results. Furthermore, we precondition that an LBS is able to dynamically negotiate with the MT the way position updates are sent. For this purpose different update strategies are possible (compare [LeRo02], [Kuep05]). Among them are the periodic (updates are sent in fixed intervals), zone-based (an update is sent whenever an MT enters or leaves a given zone), or distance-based strategy (an update is sent whenever the distance between the previously reported update and the current update exceeds a certain threshold). An architectural sketch of how these strategies could be realized on MTs is given in [KuTr05] and [CCR03]. PROCEEDINGS OF THE 2nd WORKSHOP ON POSITIONING, NAVIGATION AND COMMUNICATION (WPNC’05) & 1st ULTRA-WIDEBAND EXPERT TALK (UET'05) 165