Research on Fuzzing Testing Technology Based on Dynamic Taint Analysis of Industrial Control Software

• 1、Software Security Center,Beijing University of Posts and Telecommunications,Beijing 100876

Abstract：In recent years, with the increasingly severe security situation of industrial control system, industrial control software security problem is particularly prominent. The fundamental solution to solve security problems of Industrial control software lies in discovering security vulnerabilities and fixes. In this paper, the security testing of industrial control software is concentrated on. Firstly, this paper analyzes the relevant characteristics of industrial control software and common fuzzing technology. The existing fuzzy testing technology has the disadvantage of low efficiency and other issues. From the perspective of data variability position selection, combined with industrial control software features, this paper proposed a fuzzing method based on dynamic taint analysis of industrial control software, which optimized the existing method both in the storage and analysis of the tainted data. This method is mainly through the industrial dynamic taint analysis software to assist in the completion of the selection of the fuzzing position variation data, thereby reducing invalid test cases of the fuzzing test, and improve the efficiency of fuzzing test. Finally, this paper conducted several experiments with the existing test environment. The test results show that this method significantly improve the detection efficiency compared with the existing fuzzing method. Furthermore, this experiment successfully found some cases of denial of service vulnerability, which strongly proves the correctness of this method.

Keywords： industrial control software dynamic taint analysis fuzzing test