Human mobility tracks as FAIR data: Designing a privacy-preserving repository for GNSS-based activity tracking data

Abstract

Abstract. Devices with integrated global navigation satellite system (GNSS) receivers have enabled citizens to accurately record activities such as bicycle trips, runs, and walks. Due to its spatiotemporal extent and high level of detail, GNSS-based activity tracking data is a valuable source of information on active modes of transportation. At the same time, movement recordings of individuals are sensitive data and are associated with privacy concerns. In this work, we present a privacy-aware platform where citizens can contribute GNSS tracks to an open repository. The repository is published according to the FAIR data principles: findable, accessible, interoperable, and reusable. This provides the opportunity to use the data as a benchmark for the development of GNSS trajectory processing methods. The platform’s privacy module processes each track before publication, concealing stay points, generalizing the tracks in the temporal dimension, and suppressing tracks in sparsely populated areas. This approach mitigates the most likely re-identification attacks and limits the amount of information that could leak if an attacker succeeds with re-identification. As a residual risk remains, the platform sensitizes users to privacy risks and enables them to make informed decisions about publishing their data.