Detection of global, metamorphic malware variants using control and data flow analysis

摘要: Current malware detection and classification tools fail to adequately address variants that are generated automatically using new polymorphic metamorphic transformation engines can produce bear no resemblance one another. approaches this problem by employing syntactic signatures mimic the underlying control structures such as call- flow-graphs. These techniques, however, easily defeated program diversification techniques. This hampers our ability defend against zero day attacks perpetrated auto “replicating”, rapidly spreading variants. In paper, we present a form of abstract signature generation is based on extracting semantic summaries code immune most transformations. We also results initial, experimental evaluation proposed approach.