EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

RGANet: A Human Activity Recognition Model for Extracting Temporal and Spatial Features from WiFi Channel State Information

Jianyuan Hu, Fei Ge iD, Xinyu Cao, Zhimin Yang

Sensors · 03.02.2025 · 25 (3) · 918

DOI: 10.3390/s25030918 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 03.02.2025. Verfasst von Jianyuan Hu und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science, Central China Normal University, Wuhan 430070, China. Hauptthemen laut MEDLINE: Wireless Technology, Human Activities. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, China Higher Education Institution Industry-University-Research Innovation Fund.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the rapid advancement of communication technologies, wireless networks have not only transformed people’s lifestyles but also spurred the development of numerous emerging applications and services. Against this backdrop, research on Wi-Fi-based human activity recognition (HAR) has become a hot topic in both academia and industry. Channel State Information (CSI) contains rich spatiotemporal information. However, existing deep learning methods for human activity recognition (HAR) typically focus on either temporal or spatial features. While some approaches do combine both types of features, they often emphasize temporal sequences and underutilize spatial information. In contrast, this paper proposes an enhanced approach by modifying residual networks (ResNet) instead of using simple CNN. This modification allows for effective spatial feature extraction while preserving temporal information. The extracted spatial features are then fed into a modifying GRU model for temporal sequence learning. Our model achieves an accuracy of 99.4% on the UT_HAR dataset and 99.24% on the NTU-FI HAR dataset. Compared to other existing models, RGANet shows improvements of 1.21% on the UT_HAR dataset and 0.38% on the NTU-FI HAR dataset.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Wireless TechnologyHuman ActivitiesHumansAlgorithmsNeural Networks, ComputerDeep Learning

Stichwörter der Autor:innen

Channel State Information (CSI)Deep Learning (DL)Human Activity Recognition (HAR)

Autor:innen und Institutionen

  • School of Computer Science, Central China Normal University, Wuhan 430070, China.

ORCID-Profile: Fei Ge

Förderung

  • National Natural Science Foundation of China
  • China Higher Education Institution Industry-University-Research Innovation Fund

Zitieren

APA

Hu, J., Ge, F., Cao, X., & Yang, Z. (2025). RGANet: A Human Activity Recognition Model for Extracting Temporal and Spatial Features from WiFi Channel State Information. Sensors, 25(3), 918. https://doi.org/10.3390/s25030918

Vancouver

Hu J, Ge F, Cao X, Yang Z. RGANet: A Human Activity Recognition Model for Extracting Temporal and Spatial Features from WiFi Channel State Information. Sensors. 2025;25(3):918. doi:10.3390/s25030918

BibTeX
@article{Hu2025,
  author = {Hu, Jianyuan and Ge, Fei and Cao, Xinyu and Yang, Zhimin},
  title = {RGANet: A Human Activity Recognition Model for Extracting Temporal and Spatial Features from WiFi Channel State Information},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {3},
  pages = {918},
  doi = {10.3390/s25030918}
}
Kostenlos starten →