Energy-efficient Position Tracking via Trajectory Modeling

Abstract

In recent years, the market of location-aware applications has become increasingly larger. However, one of the main problems is still partially unsolved: battery consumption. The low operating time of the batteries hampers the opportunities for collecting users' location data in order to provide a variety of services,ranging from events suggestion to sport tracking. This article proposes a way to diminish the battery consumption by reducing the communication between the clients and the server. The approach is to model the trajectory of the users, so that the clients communicate a new position only when necessary: namely,when the predicted position excessively deviates from the actual one. In order to demonstrate the idea, a prototype has been implemented. This paper evaluates the performance of the model-based technique and it compares the proposed approach with the approach commonly used for location tracking in mobile devices.