Demystify the Fuzzing Methods: A Comprehensive Survey

Sanoop Mallissery, Yu Sung Wu*

*此作品的通信作者

研究成果: Article同行評審

16 引文 斯高帕斯(Scopus)

摘要

Massive software applications possess complex data structures or parse complex data structures; in such cases, vulnerabilities in the software become inevitable. The vulnerabilities are the source of cyber-security threats, and discovering this before the software deployment is challenging. Fuzzing is a vulnerability discovery solution that resonates with random-mutation, feedback-driven, coverage-guided, constraint-guided, seed-scheduling, and target-oriented strategies. Each technique is wrapped beneath the black-, white-, and grey-box fuzzers to uncover diverse vulnerabilities. It consists of methods such as identifying structural information about the test cases to detect security vulnerabilities, symbolic and concrete program states to explore the unexplored locations, and full semantics of code coverage to create new test cases. We methodically examine each kind of fuzzers and contemporary fuzzers with a profound observation that addresses various research questions and systematically reviews and analyze the gaps and their solutions. Our survey comprised the recent related works on fuzzing techniques to demystify the fuzzing methods concerning the application domains and the target that, in turn, achieves higher code coverage and sound vulnerability detection.

原文English
文章編號3623375
期刊ACM Computing Surveys
56
發行號3
DOIs
出版狀態Published - 31 3月 2024

指紋

深入研究「Demystify the Fuzzing Methods: A Comprehensive Survey」主題。共同形成了獨特的指紋。

引用此