The exponential growth of Internet of Things (IoT) devices has brought about transformative advancements across industries but has also introduced significant cybersecurity challenges. Traditional centralized threat detection methods often fall short in meeting the scalability, heterogeneity, and privacy demands of IoT networks. To address these limitations, this paper presents a Privacy-Preserving Federated Learning (PPFL) framework designed to deliver secure and efficient threat detection in IoT environments. By leveraging federated learning, the PPFL framework enables decentralized data processing, reducing privacy risks while improving detection accuracy through the use of diverse and distributed data sources. Comprehensive evaluations conducted using the UNSWNB15 dataset reveal that the PPFL framework outperforms traditional approaches in terms of accuracy, latency, scalability, and privacy preservation. These results highlight the potential of PPFL as a robust solution to enhance IoT network security and pave the way for future innovations in privacy-preserving cybersecurity methodologies.
Atoum et al. (Mon,) studied this question.