An industrial control system malicious sample generation method based on adversarial learning

Abstract

The invention discloses an industrial control system malicious sample generation method based on adversarial learning. According to the method, a confrontation sample of an industrial control intrusion detection system based on a machine learning method is calculated by utilizing an optimization solution through a confrontation learning technology. Attack samples which can be originally found by the intrusion detection system can generate corresponding new countermeasure samples after being processed by the method. The confrontation sample can bypass an original intrusion detector (identifiedas normal), and the attack effect is still kept. By actively generating the malicious samples of the industrial control system, the safety of the industrial control system is effectively guaranteed, and preventive measures are taken.

Description

technical field [0001] The invention belongs to the field of industrial control system security, and relates to a method for generating malicious samples of an industrial control system based on confrontation learning. Background technique [0002] The frequent occurrence of industrial control system security incidents has attracted great attention from countries all over the world. In order to actively respond to industrial control security issues, various countermeasures have been launched, such as the formulation of normative standards and the deployment of intrusion detection. In industrial control systems, network-based intrusion detectors are very typical and common. By deploying a network anomaly detection device between the application network (IT) and the control network (OT), the security of the industrial control system can be greatly improved. At present, machine learning-based anomaly detection algorithms can achieve better performance in identifying abnormal ...

AI Technical Summary

Problems solved by technology

[0004] For the intrusion detection system of industrial control system based on machine learning algorithm, if the attacker deliberately resists the attack, it will not be able to identify the malicious traffic that can be identified normally.

This will bring great security risks to the industrial control system.

At present, the existing industrial control security intrusion detection research lacks research on the part of adversarial learning, and it is very meaningful to simulate attackers to actively generate adversarial samples for industrial control machine learning detection algorithms

Images