Anomaly Detection in IoT Networks Using Federated Machine Learning Approaches
R. Vidhya, D. Lognathan, Sibyala Saranya, P.N. Periyasamy, S. Sumathi · International Journal of Computational and Experimental Science and Engineering · 2025
The rapid growth of Internet of Things (IoT) networks has brought forth new challenges in ensuring the security and reliability of devices and data. Anomaly detection in IoT networks is crucial for identifying malicious activities, faulty devices, and abnormal behaviors that could lead to system failures or security breaches. Traditional centralized machine learning models for anomaly detection require the aggregation of sensitive data from multiple IoT devices, raising concerns about privacy and scalability. To address these challenges, this paper proposes a federated machine learning (FML) approach for anomaly detection in IoT networks. Federated learning allows models to be trained locally on devices without sharing raw data, thus preserving privacy while leveraging the collective knowledge of decentralized devices. The proposed approach integrates anomaly detection algorithms with federated learning frameworks to identify network anomalies while maintaining data confidentiality. Experimental results demonstrate that the federated learning-based anomaly detection model achieves high detection accuracy, reduces communication overhead, and scales effectively across diverse IoT devices. This approach offers a promising solution for real-time security monitoring in large-scale IoT environments, where data privacy and resource efficiency are paramount.