Fine-grained Stop-Move Detection in UWB-based Trajectories

Ultra-wideband (UWB) localization enables user tracking with high spatio-temporal resolution, whose exploitation for detecting higher-level mobility patterns is largely unexplored. We study whether i) existing detection techniques, developed for coarser-grained localization, apply also to UWB trajectories, and ii) the quantitative extent to which this enables finer-grained analyses. We focus on the well-known stop-move pattern, and offer a concrete use case of capturing visits in a real museum. We contribute a novel metric suited to the high UWB spatio-temporal resolution and use it to evaluate representative techniques. We deploy a UWB system in a 25 x15 m(2) museum area and base our analysis on 70000+ positions and 200+ ground-truth stops. These are very close in space and time, yet results confirm very accurate spatio-temporal estimation in the vast majority of cases.