SAS: Semantics Aware Signature Generation for Polymorphic Worm Detection
Deguang Kong,Yoon-Chan Jhi,Tao Gong,Sencun Zhu,Peng Liu,Hongsheng Xi
DOI: https://doi.org/10.1007/s10207-011-0132-7
2010-01-01
Abstract:String extraction and matching techniques have been widely used in generating signatures for worm detection, but how to generate effective worm signatures in an adversarial environment still remains a challenging problem. For example, attackers can freely manipulate byte distributions within the attack payloads and thus inject well-crafted noisy packets to contaminate the suspicious flow pool. To address these attacks, we propose SAS, a novel S emantics A ware S tatistical algorithm for automatic signature generation. When SAS processes packets in a suspicious flow pool, it uses data flow analysis techniques to remove non-critical bytes. We then apply a hidden Markov model (HMM) to the refined data to generate state-transition-graph-based signatures. To our best knowledge, this is the first work combining semantic analysis with statistical analysis to automatically generate worm signatures. Our experiments show that the proposed technique can accurately detect worms with concise signatures. Moreover, our results indicate that SAS is more robust to the byte distribution changes and noise injection attacks compared to Polygraph and Hamsa.