Vulnerability mining method based on improved generative adversarial network framework

Abstract

The invention discloses a vulnerability mining method based on an improved generative adversarial network framework, provides an improved generative adversarial network model framework, and solves the problems of poor target function training effect in a test case generation process and mode collapse in a model training process, too slow convergence speed and not high enough quality of the generated test case due to the fact that feedback information obtained by the generator network from the discriminator network is insufficient in the training process. The method enables the model to achieve faster convergence and obtain higher quality test cases. and finally, a vulnerability mining system is designed and realized based on a Modbus-TCP protocol, and a fuzzy test is performed in a simulation environment and a real industrial environment . According to the method, the receiving rate of the test case is improved, various abnormities of the tested target can be triggered, and vulnerabilities of the Modbus-TCP protocol can be found, so that the problems of low receiving rate and low vulnerability mining capability in a traditional vulnerability mining method are solved.

Description

technical field [0001] The invention relates to the network security field of an industrial control system, in particular to a test case generation method based on an improved generation confrontation network framework and a vulnerability mining method based on the Modbus-TCP protocol. Background technique [0002] Industrial control systems are widely used in infrastructure industries that affect people's lives, such as electric power, petroleum, steel, and rail transit. They are promoting industrial upgrading in traditional industries, accelerating transformation, and rapidly improving social productivity. The production and operation process of the manufacturing industry will be controlled by information technology. At the same time of automatic control, it also forces itself to interact more frequently with the outside world, which gradually increases the "exposure" of the originally relatively closed environment, resulting in a greatly increased security risk. [0003] ...

AI Technical Summary

Problems solved by technology

In recent years, there has been some progress in the research of vulnerability mining technology at home and abroad, but the problems of low acceptance rate and low vulnerability mining rate have appeared in the process of applying the test case construction method of industrial control protocol to the vulnerability mining experiment.

Images