Enhancing Cyber-Physical System Security: A Novel Approach to Real-Time Cyber Attack Detection and Mitigation

Cyber-Physical Systems (CPS) security has become crucial with the rapid growth of Internet of Things (IoT) devices in different sectors, including critical infrastructure and industrial systems. Cyber-attacks directed at CPS may result in serious repercussions, such as bodily harm, monetary loss, or even fatality. As a result, in IoT-enabled CPS systems, there is an urgent need for reliable and effective detection mechanisms to recognize and neutralize cyber threats. This research proposes an integrated system for cyberattack detection in IoT-enabled CPS. The suggested system efficiently detects and addresses cyber threats by combining an anomaly detection technique with a decision tree. The CPS environment’s Internet of Things (IoT) devices constantly produce data streams on network traffic, sensor readings, and system operations. Gathering and preprocessing these data streams is to extract pertinent information for analysis. The system integrates principal component analysis (PCA) with deep neural networks (DNN) for intrusion detection and network traffic analysis to monitor and analyze communication patterns between IoT devices, servers, and external networks. At the network level, this aids in the identification of possible attack pathways and malicious activity. When a cyber-attack or anomaly is detected, the system initiates the proper reaction processes to lessen the effects and stop additional harm. It could entail notifying system administrators for human intervention, changing access rules, and isolating hacked devices. Finally, the proposed approach shows the significant impact of detecting cyberattacks from networking datasets.