A prototype of event-based tracking system for mobile users

Abstract

This paper describes prototype development of event-based tracking for mobile users. Nowadays, queries about location and activities are the most common question people send to other people in their social connections. This phenomenon has highlighted the need to provide a service to monitor positions and physical activities of multiple users. There are several challenges that need to be dealt with when developing the service of tracking system. In this paper a framework that facilitates service to monitor location and user physical activities is developed. A mechanism to reduce periodic update of position and sensor data for activity recognition is proposed for the mobile side. It is done by estimating degree of temporal sensor data changes as the trigger for the updating process. In the server a simple yet robust classification algorithm is applied and the accuracy of activity recognition is also evaluated. The framework prototype has also been implemented in real scenarios. The results of experiments show a promising results with 95% recognition rate for cycling activities and 80% for both walking and running activities.