EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Stability-Controlled Continual Federated Learning for Energy-Harvesting AIoT Systems

Junsoo Park iD, Ikjune Yoon iD, Dong Kun Noh iD

Sensors · 23.05.2026 · 26 (11) · 3325

DOI: 10.3390/s26113325 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 23.05.2026. Verfasst von Junsoo Park und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Biomedical Systems, Soongsil University, Seoul 06978, Republic of Korea. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Institute of Information & Communications Technology Planning & Evaluation, Institute of Information & Communications Technology Planning & Evaluation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Energy-harvesting (EH) AIoT systems enable long-term autonomous operation but suffer from time-varying energy availability, which makes stable learning difficult. In such environments, federated learning (FL) is prone to energy depletion (blackout), while continual learning is required to handle evolving data distributions, leading to a trade-off between energy stability and catastrophic forgetting. In this paper, we propose a stability-controlled continual federated learning framework that jointly regulates local training intensity and rehearsal usage based on the residual energy state. The proposed method is derived from a Lyapunov drift-plus-penalty formulation and implemented as a lightweight mode-based control policy. Simulation results using real solar energy traces show that the proposed method significantly reduces blackout while improving accuracy and mitigating forgetting compared to existing approaches. These results demonstrate the effectiveness of energy-aware joint control for stable continual federated learning in EH-AIoT systems.

Schlagwörter

Stichwörter der Autor:innen

Lyapunov stabilitycontinual learningenergy-aware controlenergy-harvesting AIoTfederated learning

Autor:innen und Institutionen

  • School of Biomedical Systems, Soongsil University, Seoul 06978, Republic of Korea.
  • Division of AI Computer Science & Engineering, Kyonggi University, Suwon 16227, Republic of Korea.
  • School of AI Convergence, Soongsil University, Seoul 06978, Republic of Korea.

ORCID-Profile: Junsoo Park, Ikjune Yoon, Dong Kun Noh

Förderung

  • Institute of Information & Communications Technology Planning & Evaluation
  • Institute of Information & Communications Technology Planning & Evaluation

Zitieren

APA

Park, J., Yoon, I., & Noh, D. K. (2026). Stability-Controlled Continual Federated Learning for Energy-Harvesting AIoT Systems. Sensors, 26(11), 3325. https://doi.org/10.3390/s26113325

Vancouver

Park J, Yoon I, Noh DK. Stability-Controlled Continual Federated Learning for Energy-Harvesting AIoT Systems. Sensors. 2026;26(11):3325. doi:10.3390/s26113325

BibTeX
@article{Park2026,
  author = {Park, Junsoo and Yoon, Ikjune and Noh, Dong Kun},
  title = {Stability-Controlled Continual Federated Learning for Energy-Harvesting AIoT Systems},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {11},
  pages = {3325},
  doi = {10.3390/s26113325}
}
Kostenlos starten →