Step Detection Enhanced by Anomaly Filtering
Sayyaf MI, Zhu N, Renaudin V, Feigl T (2025)
Publication Type: Conference contribution
Publication year: 2025
Publisher: IEEE
City/Town: New York City
Pages Range: 223-226
Conference Proceedings Title: 2025 IEEE Applied Sensing Conference (APSCON)
DOI: 10.1109/APSCON63569.2025.11144204
Abstract
This paper presents a novel anomaly detection algorithm that improves the accuracy of step detection by eliminating false detections caused by signals that mimic walking, such as hand movements, vibrations, dropping the phone, rapid movements, and actions such as standing up or sitting down, which can affect reliability. Numerous anomalous movements can mimic walking, making it difficult to identify and rule them all out. This challenge underscores the need for robust anomaly detection methods that are able to distinguish true walking signals from a wide range of non-walking movements with low uncertainty. The proposed algorithm effectively distinguishes normal walking from such anomalies and achieves a high detection rate of 95.2% ± 0.7% for normal movements in a 40 km long dataset recorded from eight people using a smartphone in texting, pocket, and swing mode. The algorithm detects anomalous movements with a detection rate of 82.5% ± 1.25% and thus successfully filters out these false signals. Furthermore, the suitability of the algorithm for mobile devices is emphasized by evaluating its computational and energy efficiency using metrics such as floating point operations per second (FLOPs), CPU processing time and energy consumption.
Authors with CRIS profile
Involved external institutions
Université Gustave Eiffel
France (FR)
Fraunhofer-Institut für Integrierte Schaltungen (IIS)
Germany (DE)
France (FR)
Fraunhofer-Institut für Integrierte Schaltungen (IIS)
Germany (DE)
How to cite
APA:
Sayyaf, M.I., Zhu, N., Renaudin, V., & Feigl, T. (2025). Step Detection Enhanced by Anomaly Filtering. In 2025 IEEE Applied Sensing Conference (APSCON) (pp. 223-226). New York City: IEEE.
MLA:
Sayyaf, Mohamad Issam, et al. "Step Detection Enhanced by Anomaly Filtering." Proceedings of the 2025 IEEE Applied Sensing Conference (APSCON) New York City: IEEE, 2025. 223-226.
BibTeX: Download