EPCAD: Efficient and Privacy-Preserving Data Anomaly Detection Scheme for Industrial Control System Networks
Hanjun Gao, Liang Yuan, Fei Yin, Gang Shen · Journal of Physics Conference Series · 2021
Abstract With the integration of Internet and industry, traditional industrial control system (ICS) has faced cyber-security risks and challenges due to interacting with the Internet. In this paper, we propose an efficient and privacy-preserving data anomaly detection scheme (EPCAD) for ICS. The scheme, a combination of a homomorphic cryptosystem and the support vector machine (SVM) algorithm, has the capability of efficiently detecting anomalies in data without compromising data information. Security analysis result shows that the EPCAD scheme has the following advantages: protect the data in the programmable logic controller (PLC); ensure that the classification parameters of anomaly detection server (ADS) are not compromised. Performance evaluation analysis demonstrates that the EPCAD scheme has significant advantages in terms of computational costs and communication overheads.