EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TCN-MAML: A TCN-Based Model with Model-Agnostic Meta-Learning for Cross-Subject Human Activity Recognition

Chih-Yang Lin iD, Chia-Yu Lin, Yu-Tso Liu iD, Yi-Wei Chen, Hui-Fuang Ng iD, Timothy K Shih

Sensors · 2025

DOI: 10.3390/s25134216 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Chih-Yang Lin und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mechanical Engineering, National Central University, Taoyuan City 32001, Taiwan. Hauptthemen laut MEDLINE: Human Activities. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science and Technology Council.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Human activity recognition (HAR) using Wi-Fi-based sensing has emerged as a powerful, non-intrusive solution for monitoring human behavior in smart environments. Unlike wearable sensor systems that require user compliance, Wi-Fi channel state information (CSI) enables device-free recognition by capturing variations in signal propagation caused by human motion. This makes Wi-Fi sensing highly attractive for ambient healthcare, security, and elderly care applications. However, real-world deployment faces two major challenges: (1) significant cross-subject signal variability due to physical and behavioral differences among individuals, and (2) limited labeled data, which restricts model generalization. To address these sensor-related challenges, we propose TCN-MAML, a novel framework that integrates temporal convolutional networks (TCN) with model-agnostic meta-learning (MAML) for efficient cross-subject adaptation in data-scarce conditions. We evaluate our approach on a public Wi-Fi CSI dataset using a strict cross-subject protocol, where training and testing subjects do not overlap. The proposed TCN-MAML achieves 99.6% accuracy, demonstrating superior generalization and efficiency over baseline methods. Experimental results confirm the framework’s suitability for low-power, real-time HAR systems embedded in IoT sensor networks.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Human ActivitiesHumansWearable Electronic DevicesNeural Networks, ComputerWireless TechnologyAlgorithmsMachine Learning

Stichwörter der Autor:innen

MAMLTCNhuman activity recognitionwireless sensor networks

Autor:innen und Institutionen

  • Department of Mechanical Engineering, National Central University, Taoyuan City 32001, Taiwan.
  • Department of Computer Science and Information Engineering, National Central University, Taoyuan City 32001, Taiwan.
  • Department of Computer Science, Universiti Tunku Abdul Rahman, Kampar 31900, Perak, Malaysia.

ORCID-Profile: Chih-Yang Lin, Yu-Tso Liu, Hui-Fuang Ng

Förderung

  • National Science and Technology Council

Zitieren

APA

Lin, C. Y., Lin, C. Y., Liu, Y. T., Chen, Y. W., Ng, H. F., & Shih, T. K. (2025). TCN-MAML: A TCN-Based Model with Model-Agnostic Meta-Learning for Cross-Subject Human Activity Recognition. Sensors. https://doi.org/10.3390/s25134216

Vancouver

Lin CY, Lin CY, Liu YT, Chen YW, Ng HF, Shih TK. TCN-MAML: A TCN-Based Model with Model-Agnostic Meta-Learning for Cross-Subject Human Activity Recognition. Sensors. 2025. doi:10.3390/s25134216

BibTeX
@article{Lin2025,
  author = {Lin, Chih-Yang and Lin, Chia-Yu and Liu, Yu-Tso and Chen, Yi-Wei and Ng, Hui-Fuang and Shih, Timothy K},
  title = {TCN-MAML: A TCN-Based Model with Model-Agnostic Meta-Learning for Cross-Subject Human Activity Recognition},
  journal = {Sensors},
  year = {2025},
  doi = {10.3390/s25134216}
}
Kostenlos starten →