Federated Learning for Privacy-Preserving Intrusion Detection in Heterogeneous IOT Networks
International Journal of Science, Strategic Management and Technology · 18.03.2026 · 02 (03) · 1-9
DOI: 10.55041/ijsmt.v2i3.124 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Science, Strategic Management and Technology, online erschienen am 18.03.2026. Verfasst von Jay Shrimali.
Abstract (Originalsprache)
The rapid proliferation of Internet of Things (IoT) devices across industrial, domestic, and healthcare environments has created significant challenges in network security. Traditional centralised intrusion detection systems (IDS) are inadequate for IoT deployments due to privacy constraints, communication overhead, and the heterogeneous nature of device ecosystems. This paper proposes a federated learning-based intrusion detection framework (FL-IDS) that enables distributed model training across IoT edge nodes without transmitting raw data to a central server. The proposed architecture employs a lightweight convolutional-recurrent neural network (CRNN) at each participating node, with a modified FedAvg aggregation strategy that accounts for data imbalance and device heterogeneity. Experiments conducted on the N-BaIoT and TON-IoT benchmark datasets demonstrate that FL-IDS achieves a detection accuracy of 97.43% and an F1-score of 96.89%, while reducing communication overhead by 61.2% compared to centralised approaches. The framework also demonstrates robustness against Byzantine attacks and adversarial data poisoning, making it suitable for deployment in real-world constrained environments
Zitieren
APA
Shrimali, J. (2026). Federated Learning for Privacy-Preserving Intrusion Detection in Heterogeneous IOT Networks. International Journal of Science, Strategic Management and Technology, 02(03), 1-9. https://doi.org/10.55041/ijsmt.v2i3.124
Vancouver
Shrimali J. Federated Learning for Privacy-Preserving Intrusion Detection in Heterogeneous IOT Networks. International Journal of Science, Strategic Management and Technology. 2026;02(03):1-9. doi:10.55041/ijsmt.v2i3.124
BibTeX
@article{Shrimali2026,
author = {Shrimali, Jay},
title = {Federated Learning for Privacy-Preserving Intrusion Detection in Heterogeneous IOT Networks},
journal = {International Journal of Science, Strategic Management and Technology},
year = {2026},
volume = {02},
number = {03},
pages = {1--9},
doi = {10.55041/ijsmt.v2i3.124}
}
EUVIMED