Implementation and Validation on Improvements for Dynamic Analysis System to Detect Evasive Mobile Malware

Abstract

As the use of mobile devices such as smart phones and tablets has increased, and with a heavier reliance on
them in everyday life on the rise, the malicious codes targeting mobile devices have been steadily increasing. A
worldwide effort has been directed towards developing technologies to detect and cope with mobile malicious apps.
However, malicious apps are also becoming more intelligent and sophisticated to increase survivability by bypassing the
existing analysis means, obfuscating code, detecting the analysis environment (a sandbox, emulator, etc), which is used
to analyze malicious apps, and suspending malicious behavior accordingly. In this paper, we examine the analysisevasive techniques used in malicious mobile apps, and the methods to deal with them. We also implement and verify the
improvement plans for a dynamic analysis system to incapacitate the analysis-evasive behaviors of malicious apps.