Detection and Attribution of Cyber Attacks in IOT Enabled Cyber-Physical-Systems

The rapid evolution of 5G technology and the widespread integration of Internet of Things (IoT) devices in Cyber-Physical Systems (CPS) have introduced significant secu- rity challenges. Traditional intrusion detection systems struggle to identify sophisticated and zero-day cyber-attacks in such dynamic and complex environments. This project, titled Detection and Attribution of Cyber Attacks in IoT-Enabled Cyber-Physical Systems, proposes an intelligent and adaptive Network Intrusion Detection System (NIDS) enhanced by Generative Adversarial Networks (GANs). GANs are used to generate realistic syn- thetic attack data, which helps to address the issues of data scarcity and imbalance in existing datasets. A deep learning- based model is trained on this enriched data to accurately detect and classify various types of intrusions in real-time. The system is integrated with a user-friendly web interface using Flask, making it accessible for real-time monitoring and prediction. Testing on benchmark datasets like CICIDS2017 and NSL-KDD demonstrates improved performance in terms of accuracy, recall, and precision. The proposed solution ensures scalability, real-time detection, and adaptability, making it highly suitable for securing next-generation 5G and IoT-based infrastructures