A New Trend for the Blue Team - Using a Practical Symbolic Engine to Detect Evasive Forms of Malware/Ransomware
time
02:10 ~ 02:55
site
R1
Blue Teams and anyone on the defensive side face various challenges when it comes to reverse engineering suspected malware or ransomware binaries, especially ones with obfuscation techniques such as variants, embedded exploits and complex ransomware. First, identifying whether the sample is even worth the effort (what makes it unique/challenging/new), and second, choosing either static, dynamic analysis, or both! With static analysis, you give up the ability to detect obfuscated malicious programs only visible during run-time, and dynamic analysis is both labor and time intensive, and requires a high-degree of skill and experience, not to mention the threat of the binary escaping your sandbox emulation or virtualization environment.
We believe there may be a new tool in the Blue Team’s toolbox, through the use of a symbolic execution engine to detect and analyze suspected malware/ransomware binaries. A practical symbolic engine can help by parsing through many of the possible execution paths of the binary, and having these pathways represented as symbols. This engine can help provide malicious execution paths analysis with relatively low computing resources, analyze contextual relationships based on instruction semantics, taint and fuzzy identification of obfuscated APIs.
Using our practical symbolic engine based on the combination and improvement of academic and practical research, you can identify and detect various exploit, techniques, and multiple malware/ransomware variants via symbolic signature attack techniques and ransomware behaviors in a fully static situation. Even if the malware binary is obfuscated, we can still statically analyze it and detect it effectively.
ShengHao Ma
Hank Chen
