Abdallah Dawoud (CISPA Helmholtz Center for Information Security), Sven Bugiel (CISPA Helmholtz Center for Information Security)

Android's application framework plays a crucial part in protecting users' private data and the system integrity. Consequently, it has been the target of various prior works that analyzed its security policy and enforcement. Those works uncovered different security problems, including incomplete documentation, permission re-delegation within the framework, and inconsistencies in access control. However, all but one of those prior works were based on static code analysis. Thus, their results provide a one-sided view that inherits the limitations and drawbacks of applying static analysis to the vast, complex code base of the application framework. Even more, the performances of different security applications---including malware classification and least-privileged apps---depend on those analysis results, but those applications are currently tarnished by imprecise and incomplete results as a consequence of this imbalanced analysis methodology. To complement and refine this methodology and consequently improve the applications that are dependent on it, we add dynamic analysis of the application framework to the current research landscape and demonstrate the necessity of this move for improving the quality of prior results and advancing the field. Applying our solution, called Dynamo, to four prominent use-cases from the literature and taking a synoptical view on the results, we verify but also refute and extend the existing results of prior static analysis solutions. From the manual investigation of the root causes of discrepancies between results, we draw new insights and expert knowledge that can be valuable in improving both static and dynamic testing of the application framework.

View More Papers

Emilia: Catching Iago in Legacy Code

Rongzhen Cui (University of Toronto), Lianying Zhao (Carleton University), David Lie (University of Toronto)

Read More

(Short) WIP: Deployability Improvement, Stealthiness User Study, and Safety...

Takami Sato, Junjie Shen, Ningfei Wang (UC Irvine), Yunhan Jia (ByteDance), Xue Lin (Northeastern University), and Qi Alfred Chen (UC Irvine)

Read More

KUBO: Precise and Scalable Detection of User-triggerable Undefined Behavior...

Changming Liu (Northeastern University), Yaohui Chen (Facebook Inc.), Long Lu (Northeastern University)

Read More

Practical Blind Membership Inference Attack via Differential Comparisons

Bo Hui (The Johns Hopkins University), Yuchen Yang (The Johns Hopkins University), Haolin Yuan (The Johns Hopkins University), Philippe Burlina (The Johns Hopkins University Applied Physics Laboratory), Neil Zhenqiang Gong (Duke University), Yinzhi Cao (The Johns Hopkins University)

Read More