Real-Time Privacy-Preserving Threat Detection in IoT Environments using Federated Learning and Differential Privacy

Motivated by the need for real-time, privacy-preserving threat detection in Internet of Things (IoT) environments, this Design Science Research develops a conceptual framework integrating federated learning and differential privacy. The proposed artifact enables collaborative threat detection across IoT devices while ensuring data privacy. Addressing key IoT challenges like limited computational resources and device heterogeneity, the research evaluates the artifact's feasibility within existing IoT security architectures, focusing on balancing privacy, detection accuracy, and system performance. Contributions include advancing theoretical knowledge and providing practical, privacy-preserving solutions for securing sensitive data in connected IoT ecosystems.