EUVIMEDObjevte EUVIMED
Open access · CC BY 4.0

DynamicFU: Contribution-Aware Dynamic Federated Unlearning for Industrial IoT

Ziang Wu iD, Buzhen He iD, Zhiwei Si iD, Xiuheng Liao iD, Chunhua Su iD

Sensors · 2026-06-11 · 26 (12) · 3714

DOI: 10.3390/s26123714 ↗

At a glance

Journal article in Sensors, published online on 2026-06-11. Written by Ziang Wu and 4 co-authors. Institution: Graduate School of Computer Science and Engineering, University of Aizu, Aizu-Wakamatsu 965-8580, Japan. 33 references. The full text is freely available (open access, CC BY 4.0). Funded by: Japan Science and Technology Agency, JST NEXUS AI Japan–Singapore project.

AbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

The Industrial Internet of Things (IIoT) increasingly relies on federated learning (FL) to enable collaborative model training without directly sharing raw traffic data across industrial sites. However, in practical IIoT deployments, clients may later request the removal of their data contributions from a trained federated model due to regulatory requirements, such as the General Data Protection Regulation (GDPR), ownership transfer, or internal data-governance policies. Such practical requirements create a strong demand for federated unlearning in IIoT applications. Furthermore, IIoT deployments often exhibit highly imbalanced client data distributions, resulting in substantially different contributions of individual clients to the global model. Nevertheless, most existing federated unlearning methods adopt a uniform unlearning strategy and fail to account for such client-level contribution gaps. To address this issue, we propose DynamicFU, a contribution-aware dynamic federated unlearning framework for IIoT deployments. The proposed method evaluates the target client from parameter-level, data-level, and performance-level perspectives and adaptively determines the unlearning strength by dynamically adjusting the number of unlearning rounds. Experimental results on public IIoT datasets show that DynamicFU substantially improves unlearning efficiency, achieving up to 22.89× speedup over Full Retrain while maintaining comparable effectiveness.

Subjects

Author keywords

Industrial Internet of Thingsfederated learningfederated unlearningprivacy-preserving

Authors and institutions

  • Graduate School of Computer Science and Engineering, University of Aizu, Aizu-Wakamatsu 965-8580, Japan.
  • School of Computer Science and Artificial Intelligence, Lanzhou University of Technology, Lanzhou 730050, China.

ORCID profiles: Ziang Wu, Buzhen He, Zhiwei Si, Xiuheng Liao, Chunhua Su

Funding

  • Japan Science and Technology Agency
  • JST NEXUS AI Japan–Singapore project

Cite

APA

Wu, Z., He, B., Si, Z., Liao, X., & Su, C. (2026). DynamicFU: Contribution-Aware Dynamic Federated Unlearning for Industrial IoT. Sensors, 26(12), 3714. https://doi.org/10.3390/s26123714

Vancouver

Wu Z, He B, Si Z, Liao X, Su C. DynamicFU: Contribution-Aware Dynamic Federated Unlearning for Industrial IoT. Sensors. 2026;26(12):3714. doi:10.3390/s26123714

BibTeX
@article{Wu2026,
  author = {Wu, Ziang and He, Buzhen and Si, Zhiwei and Liao, Xiuheng and Su, Chunhua},
  title = {DynamicFU: Contribution-Aware Dynamic Federated Unlearning for Industrial IoT},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {12},
  pages = {3714},
  doi = {10.3390/s26123714}
}
Start for free →