A system call-centric analysis and stimulation technique to automatically reconstruct Android malware behaviors

With more than 500 million of activations reported in Q3 2012, Android mobile devices are becoming ubiquitous and trends confirm this is unlikely to slow down. App stores, such as Google Play, drive the entire economy of mobile applications. Unfortunately, high turnovers and access to sensitive data have soon attracted the interests of cyber-criminals too with malware now hitting Android devices at an alarmingly rising pace. In this paper we present Copper-Droid, an approach built on top of QEMU to automatically perform out-of-the-box dynamic behavioral analysis of An-droid malware. To this end, CopperDroid presents a unified analysis to characterize low-level OS-specific and high-level Android-specific behaviors. Based on the observation that such behaviors are however achieved through the invocation of system calls, CopperDroid’s VM-based dynamic system call-centric analysis is able to faithfully describe the behav-ior of Android malware whether it is initiated from Java, JNI or native code execution. We carried out extensive experiments to assess the effec-tiveness of our analyses on a large Android malware data set of more than 1,200 samples belonging to 49 Android mal-ware families (provided by the Android Malware Genome Project) and about 400 samples over 13 families (collected from the Contagio project). Our experiments show that a proper malware stimulation strategy (e.g., sending SMS, placing calls) successfully discloses additional behaviors on a non-negligible portion of the analyzed malware samples. 1