Edge Computing Preserving Privacy Using Homomorphic Encryption Techniques for Internet of Things Networks

Edge computing enhances IoT networks by processing data closer to the source, reducing latency and bandwidth usage. However, ensuring privacy in such decentralized systems remains a challenge. Existing methods often fail to balance efficiency and security, as traditional encryption schemes introduce computational overhead and vulnerability to data breaches. To address these issues, this study proposes Privacy-Preserving Federated Learning with Homomorphic Encryption (PP-FL-HE), integrating secure aggregation with encrypted model updates to enhance privacy. The proposed method enables decentralized learning while ensuring end-to-end encryption, preventing data leakage even from untrusted edge nodes. This approach is particularly useful for IoT applications in healthcare, smart cities, and industrial automation, where sensitive data security is critical. Experimental results demonstrate that PP-FL-HE preserves model accuracy while significantly reducing computational overhead compared to existing techniques. The findings confirm the proposed method's effectiveness in enhancing security and privacy in edge-based IoT networks without compromising performance.