Energy-Efficient Rate-Adaptive Passive Traffic Sensing using smartphones

Participatory mobile sensing is becoming increasingly popular. However, the challenges to make such applications successful are multiple. The most important are the reduction of the energy consumption of the service and the automation of the sensing procedure in order to increase the user acceptance. In this paper we are focusing on vehicular traffic monitoring using embedded sensors in modern smartphones. We present ENRAPT, an Energy-Efficient Rate-Adaptive Passive Traffic Sensing system that autonomously detects the users´ context. If “Driving” is detected, a location monitoring procedure is invoked that periodically sends relevant traffic metrics to a traffic management server. We show that ENRAPT accurately detects the driving context and provides relevant information on the real time traffic conditions encountered while driving. Further, we show that ENRAPT allows to significantly extend the battery lifetime compared to other sensing schemes.