EUVIMEDEUVIMED entdecken
Crossref

AbsTrack: Multiperson Absolute Localization and Trajectory Tracking Using Commodity Wi-Fi

Shuang Zhou, Changzhen Tian, Wanming Hao, Zhaoming Lu, Celimuge Wu, Zijun Han, Ying Yang, Hao Huang

IEEE Internet of Things Journal · 2026-11 · 13 (22) · 55459-55475

DOI: 10.1109/jiot.2026.3733397 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Internet of Things Journal, 2026-11. Verfasst von Shuang Zhou und 7 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Gefördert von: China National Natural Science Foundation, Key Scientific Research in Universities of Henan Province of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the popularization of smart homes and the Internet of Things, human passive localization and tracking using Wi-Fi channel state information (CSI) have attracted much attention. However, it remains challenging to localize and track multiple people simultaneously when they move in a sensing area. While recent studies have demonstrated the potential of multiperson localization and tracking using commodity devices, they either require special equipment assistance, make Wi-Fi devices lose their basic communication function, or cannot estimate individuals’ absolute positions. To this end, we propose AbsTrack, a Wi-Fi-based system that enables multiperson absolute localization and trajectory tracking with commodity Wi-Fi devices, while addressing the key limitations of prior works. In particular, we design a multiperson absolute position estimation algorithm that performs joint angle of arrival (AoA) and time of flight (ToF) estimation at different receiver arrays. An MSE-based relative ToF (rToF) indicator is further proposed to match AoA estimates corresponding to the same individual across different receivers, thereby eliminating localized pseudopoints and enabling accurate multiperson absolute position estimation. We also design a Transformer-based multiperson trajectory tracking network with a virtual training data generation method, which can achieve fine-grained trajectory tracking of multiple people while avoiding the burden of extensive real-world data collection and annotation. Extensive experiments are conducted in three different environments, and AbsTrack can, respectively, achieve average median localization and tracking errors of 17.28 and 20.51 cm for two or three moving individuals, resulting in over a 50% error reduction compared to the state-of-the-art methods.

Förderung

  • China National Natural Science Foundation
  • Key Scientific Research in Universities of Henan Province of China

Zitieren

APA

Zhou, S., Tian, C., Hao, W., Lu, Z., Wu, C., Han, Z., Yang, Y., & Huang, H. (2026). AbsTrack: Multiperson Absolute Localization and Trajectory Tracking Using Commodity Wi-Fi. IEEE Internet of Things Journal, 13(22), 55459-55475. https://doi.org/10.1109/jiot.2026.3733397

Vancouver

Zhou S, Tian C, Hao W, Lu Z, Wu C, Han Z, et al. AbsTrack: Multiperson Absolute Localization and Trajectory Tracking Using Commodity Wi-Fi. IEEE Internet of Things Journal. 2026;13(22):55459-55475. doi:10.1109/jiot.2026.3733397

BibTeX
@article{Zhou2026,
  author = {Zhou, Shuang and Tian, Changzhen and Hao, Wanming and Lu, Zhaoming and Wu, Celimuge and Han, Zijun and Yang, Ying and Huang, Hao},
  title = {AbsTrack: Multiperson Absolute Localization and Trajectory Tracking Using Commodity Wi-Fi},
  journal = {IEEE Internet of Things Journal},
  year = {2026},
  volume = {13},
  number = {22},
  pages = {55459--55475},
  doi = {10.1109/jiot.2026.3733397}
}
Kostenlos starten →