EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

IoT Threat Detection using Federated Learning and Edge AI with Random Forest Optimization

Fortune Ifeanyi, Victor Eze

Nature Journal of Emerging Sciences Technologies and Innovations · 28.11.2025 · 3 (2) · 92-103

DOI: 10.65752/vv2vz583 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Nature Journal of Emerging Sciences Technologies and Innovations, online erschienen am 28.11.2025. Verfasst von Fortune Ifeanyi und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid increase of Internet of Things (IoT) devices has introduced significant cybersecurity risks due to their limited computational resources, making them vulnerable to threats such as malware, Distributed Denial of Service (DDoS) attacks, and unauthorized access. Traditional intrusion detection systems are often inadequate, as they struggle to provide real-time, scalable, and privacy-preserving protection in such environments. This study presents the design and evaluation of an intelligent IoT threat detection framework using Federated Learning and Edge AI with Random Forest optimization. The model was trained on the UNSW 2018 BoT-IoT dataset for hierarchical multi-label classification and integrated with TensorFlow Federated to enable decentralized training without transferring raw data. To support deployment on resource-constrained IoT devices, the model was optimized into TensorFlow Lite (TFLite) for edge inference. Experimental results demonstrated high robustness, achieving over 98.9% accuracy across attack, category, and subcategory predictions with minimal false positives, while maintaining efficiency on edge devices. The findings confirm the feasibility of deploying lightweight, privacy-preserving, and high-performance intrusion detection mechanisms on IoT networks, offering a scalable and practical solution for enhancing real-time IoT cybersecurity.

Zitieren

APA

Ifeanyi, F., & Eze, V. (2025). IoT Threat Detection using Federated Learning and Edge AI with Random Forest Optimization. Nature Journal of Emerging Sciences Technologies and Innovations, 3(2), 92-103. https://doi.org/10.65752/vv2vz583

Vancouver

Ifeanyi F, Eze V. IoT Threat Detection using Federated Learning and Edge AI with Random Forest Optimization. Nature Journal of Emerging Sciences Technologies and Innovations. 2025;3(2):92-103. doi:10.65752/vv2vz583

BibTeX
@article{Ifeanyi2025,
  author = {Ifeanyi, Fortune and Eze, Victor},
  title = {IoT Threat Detection using Federated Learning and Edge AI with Random Forest Optimization},
  journal = {Nature Journal of Emerging Sciences Technologies and Innovations},
  year = {2025},
  volume = {3},
  number = {2},
  pages = {92--103},
  doi = {10.65752/vv2vz583}
}
Kostenlos starten →