SecuSCADA: Building Secure SCADA Network with Obfuscated Malware Detection Technique

Supervisory Control and Data Acquisition system are used in various industries for monitoring and controlling industrial processes and infrastructure. They are essential for keeping these critical systems running smoothly and safely collect and analyze real-time data from various sensors, machines, and other devices to monitor and control the processes. An attack on a SCADA system can have significant and far-reaching consequences for the organization responsible for the system, as well as for public safety and the environment. It is critical for organizations to take proactive steps to protect SCADA systems from attacks, including implementing security features including network segmentation, access limits, intrusion detection, and periodic vulnerability assessments and penetration testing. In this research paper, a model is proposed to build a secured SCADA network by detecting Obfuscated malware using machine learning algorithms. The SCADA server is shielded using a honeypot in the proposed approach, the honeypot machine can detect Obfuscated malware. The proposed approach experiments with a Decision tree classifier and XGBoost algorithm for detecting such malware in the SCADA network. The XGBoost machine learning technique outperforms in comparison with the decision tree classifier and is capable of more accurately detecting this hidden malware.