论文部分内容阅读
提出了一种基于语义的恶意行为分析方法,可以对基于函数调用的攻击进行完整刻画,支持流敏感、上下文敏感且路径敏感的函数间分析。与现有方法相比可以更加准确地描述全局状态中的基于函数调用的攻击行为。针对多个恶意程序和应用程序的分析表明,该方法可以有效地识别代码中的恶意行为。
A semantic-based approach to malicious behavior analysis is proposed, which can describe the attack based on function invocation completely and support the flow-sensitive, context-sensitive and path-sensitive inter-function analysis. Compared with the existing methods can be more accurately describe the global state based on the function call attacks. Analysis of multiple malicious programs and applications shows that this method can effectively identify the malicious behavior in the code.