Intrusion detection based on traffic research and application in industrial control system

Abstract

The industrial control systems (ICS) are the combination of computer network, communication and automatic control technology. Effective defense against intrusion is the key and difficult issue of industrial control system security research. As an enterprise’s core production and operation system, industrial control system did not take into account its information security issues when it was first designed. With the rapid development of information technology and industrialization of Internet of things, these problems have gradually attracted researchers’ attention. Based on the analysis of in- dustrial control network security and the building of intrusion detection model, the paper is providing the actual industrial control systems environment, extracting the network flow and carrying on the simulation experiment. The experiment will be conducted on the NSL-KDD(the advised version of famous Intrusion Detection Data Set KDD99,used for intrusion detection experiments) and the ICS intrusion traffic dataset.Support Vector Machine (SVM) algorithm performs better in terms of testing accuracy. The performance of the precision and accuracy rate of the system will be evaluated in the experimental results.