Design of Intrusion Detection and Prevention in SCADA System for the Detection of Bias Injection Attacks
Intrusion detection and prevention system detects malicious activities that occur in the real-time SCADA systems. This system has a problem without a profound solution. The challenge of the existing intrusion detection is accuracy in the process of detecting the anomalies. In SCADA, wind turbine data are modified by the intruders and forged details are given to the server. To overcome this, the biased intrusion detection system is used for detecting the intrusion with encrypted date, time, and file location with less false-positive and false-negative rates and thereby preventing the SCADA system from further intrusion. It is done in three phases. First, Modified Grey Wolf Optimization (MGWO) is used to extract the features needed for classification and to find the best weight. Second, Entropy-based Extreme Learning Machine (EELM) is used to extort the features and detect the intruded data with its intruded time, file location, and date. Finally, the data are encrypted using the Hybrid Elliptical Curve Cryptography (HECC) to prevent further attack. Experimental results show better accuracy in both detection as well as prevention.