<![CDATA[The rapid proliferation of Internet of Things (IoT) networks has heightened the need for robust, privacy-preserving security mechanisms that ensure real-time anomaly detection. This article explores the integration of federated learning (FL) and edge computing as a promising approach to address challenges related to privacy, latency, and resource constraints in IoT environments. Employing a qualitative research methodology, this study analyzes existing literature and emerging frameworks to comprehensively assess the advantages, challenges, and future research directions of applying FL and edge computing for anomaly detection in IoT. Findings highlight that FL combined with lightweight anomaly detection algorithms deployed at the edge can significantly enhance privacy while ensuring timely intrusion detection, despite heterogeneity and limited device resources. The study suggests pathways for developing adaptive, scalable, and secure IoT networks leveraging these paradigms.]]>
Copyrights © 2025
