Key Words: IMU Privacy


  • Sensor Types: Generic, IMU, Accelerometers, IMU on Vehicle
  • Solutions: Access Control Policy


  • Location privacy is discussed in another survey, here we focus on privacy of other sensors.
  • There are many works on accelerometers-based user identification (biometric/gait) since 2005.


  • Identifying People from Gait Pattern with Accelerometers, [Biometric 2005]
    • Taxonomy: Accelerometers
    • Contributions:
      • (1) Propose that accelerometers data can be use for human identification
  • Driver Classification and Driving Style Recognition using Inertial Sensors, [IEEE IV 2013]
    • Taxonomy: IMU on Vehicle
    • Contributions:
      • (1) Show that using IMU data on the vehicle can differentiate different drivers.
  • World-driven access control for continuous sensing, [ACM CCS 2014]
    • Taxonomy: Generic, Access Control Policy
    • Contributions:
      • (1) An access control model that utilize policies and context detection methods to protect privacy automatically instead of letting the users to manully control each sensors’ permission.
      • (2) Crystallize the challenges in continuous sensing privacy.
  • Motion Sensor-based Privacy Attack on Smartphones, [arXiv 2020]
    • Taxonomy: Accelerometers
    • Contributions:
      • (1) Propose that “speech information (e.g., gender, identity) can be attacked from smartphone accelerometers”.
  • Your Tattletale Gait - Privacy Invasiveness of IMU Gait Data, [IEEE/IAPR IJCB 2020]
    • Taxonomy: IMU
    • Contributions:
      • (1) Show that IMU data can be used as a biometric by observing human movement (gait).
      • (2) Show that the location of the sensor (wrist/pocket) impact the effect of privacy invasions.
      • (3) Use an opinion survey of 566 participants to show people’s perceive on diiferent types of biometrics (e.g., weight is more important than age/gender).
  • Collecting Survey and Smartphone Sensor Data With an App, [Social Science Computer Review 2020]
    • Taxonomy: Generic
    • Contributions:
      • Come to the following conclustions through large-scale (4,300 invitee, 650 participants) survey
        • (1) People were just as willing to share such extensive digital trace data as they were in studies;
        • (2) Participants hardly differentiated between the different data requests made;
        • (3) Once participants gave consent, they did not tend to revoke it;
        • (4) Explanations regarding data collection and data usage are often not read carefully.


[Biometric 2005] Ailisto, Heikki J., Mikko Lindholm, Jani Mantyjarvi, Elena Vildjiounaite, and Satu-Marja Makela. “Identifying people from gait pattern with accelerometers.” In Biometric Technology for Human Identification II, vol. 5779, pp. 7-14. International Society for Optics and Photonics, 2005.

[IEEE IV 2013] Van Ly, Minh, Sujitha Martin, and Mohan M. Trivedi. “Driver classification and driving style recognition using inertial sensors.” In 2013 IEEE Intelligent Vehicles Symposium (IV), pp. 1040-1045. IEEE, 2013.

[ACM CCS 2014] Roesner, Franziska, David Molnar, Alexander Moshchuk, Tadayoshi Kohno, and Helen J. Wang. “World-driven access control for continuous sensing.” In Proceedings of the 2014 ACM SIGSAC Conference on Computer and Communications Security, pp. 1169-1181. 2014.

[IEEE/IAPR IJCB 2020] Rasnayaka, Sanka, and Terence Sim. “Your Tattletale Gait Privacy Invasiveness of IMU Gait Data.” In 2020 IEEE International Joint Conference on Biometrics (IJCB), pp. 1-10. IEEE.

[Social Science Computer Review 2020] Kreuter, Frauke, Georg-Christoph Haas, Florian Keusch, Sebastian Bähr, and Mark Trappmann. “Collecting survey and smartphone sensor data with an app: Opportunities and challenges around privacy and informed consent.” Social Science Computer Review 38, no. 5 (2020): 533-549.