Examining Approaches for Mobility Detection Through Smartphone Sensors
The ubiquity of smartphones with integrated positioning systems, and multiple sensors for movement detection made it possible to develop context-sensitive applications for both productivity and entertainment. Location-based games like Ingress or Pokémon Go have demonstrated the public interest in this genre of mobile-only games – games that are exclusively available for mobile devices due to their sensor integration. For these games mobility is a key component, which defines and influences the game’s flow directly.
In this paper we compare different approaches and available frameworks for mobility detection and examine the frameworks’ performances in a scenario-based evaluation.
Based on our finding we present our own approach to differentiate between different modes of public transport and other common modes of movement like walking, running or riding a bicycle. Our approach already reaches an accuracy of 87% with a small sample size.
KeywordsMobility Mobility detection Machine learning
The research presented in this paper was partially funded by the LOEWE initiative (Hessen, Germany) within the research project “Infrastruktur – Design – Gesellschaft” as project mo.de.
- 1.Schonfeld, E.: Mobile OS 2009 market share (2017). https://techcrunch.com/2010/02/23/smartphone-iphone-sales-2009-gartner/. Accessed 15 June 2018
- 2.Lau, S.L., David, K.: Movement recognition using the accelerometer in smartphones. In: Future Network and Mobile Summit (2010)Google Scholar
- 3.Wirtl, T., Nickel, C.: Aktivitätserkennung auf Smartphones. In: International Conference of the Biometrics Special Interest Group (2011)Google Scholar
- 4.Hemminki, S., Nurmi, P., Tarkoma, S.: Accelerometer-based transportation mode detection on smartphones. In: Proceedings of the 11th ACM Conference on Embedded Networked Sensor Systems (2013)Google Scholar
- 5.Anjum, A., Ilyas, M.U.: Activity recognition using smartphone sensors. In: IEEE 10th Consumer Communications and Networking Conference (2013)Google Scholar
- 6.Google: Google Awareness API (2016). https://developers.google.com/awareness/. Accessed 15 June 2018
- 7.Neura: Neura SDK (2017). https://dev.theneura.com/. Accessed 15 June 2018
- 8.Holmes, G., Donkin, A., Witten, I.H.: Weka: a machine learning workbench. In: Proceedings of the 1994 Second Australian and New Zealand Conference on Intelligent Information Systems (1994)Google Scholar