Battery-Efficient Transportation Mode Detection on Mobile Devices
Keller, Thomas; de Spindler, Alexandre (2015). Battery-Efficient Transportation Mode Detection on Mobile Devices. In: Proceedings of IEEE International Conference on Mobile Data Management (MDM 2015). Pittsburgh: IEEE. Peer reviewed.; ; ;
The ubiquitous presence of sensor-rich smartphones offers excellent opportunities to introspect individuals' contexts and activities such as the utilized modes of transport. Most of the current systems inferring transportation modes involve continuous sensing of GPS or acceleration modules, which inconveniently lead to an accelerated battery discharge. In response to this problem, this paper presents a more battery-efficient solution for the detection of transport modes. The proposed approach introduces an additional component to recognize users' mobility states, which are discriminated between stationary and moving based on cellular network information. Thus, the presented inference model enables to avoid battery-exhausting transportation mode detection when users are not travelling. The implemented system allows to distinguish the transport modes car, train and walk of six individuals with an accuracy of 95.7\\%. A test on a dataset obtained from nine individuals shows that the movement state detection achieves an accuracy of 85\\%. Furthermore, the experiment reveals that the battery lifetime can be prolonged by 75\\%.