|
Malware (such as viruses, worms, and spyware) pose a significant threat
to our computing infrastructure. There is a pressing need for techniques
for malware analysis and detection that are fast, accurate, and mostly-automatic.
In this project, we are investigating techniques for behavior-based malware detection: our algorithm focuses on detecting malicious behavior rather than searching for syntactic patterns. We specify malicious behavior in a formal language and then perform static and dynamic analyses on the code to determine whether it contains the specified behavior. This is a collaborative project, with the work in our group at UC Berkeley focussed on the computational engines (such as SAT-based decision procedures) for malware analysis and detection. PeoplePublicationsSupportThis project receives generous support from the National Science Foundation (NSF CyberTrust grant CNS-0627734), Microsoft Research, and an equipment grant from Intel. |
