Design And Implementation Of Android Application Behavior Detection System Based On Permission

To restrict an app from accessing the sensitive functionality, Android provides permission-based security model in the application framework. Usually, the users cannot tell whether these permissions are necessary or misused. How to enhance the user’s ability to control these permissions of application become the key issue to improve the reliability of Android permissions security mechanism.To solve this problem, a permission-based Android application behavior detection model was proposed.Firstly, the research status in Android security field was described.The technical background, the practical significance and the research content of this research was given too. The classification and characteristics of malicious Android application were summarized. The deficiencies and defects existing in Android permissions security mechanism were analyzed.Secondly, it summarized the analysis method and basic tools that used by the security person. On this basis, with the aid of Python, an Android application analysis and automated processing platform was built. It realized dealing with the Android application decompile, extract permissions, code injection, recompilation, resign the applications and a series of process in a automative way. A detection mechanism was proposed and designed to deal with the issue of misuse of permissions.Furthermore, based on the existing security permissions mechanism, combined with the existing research results, a fine-grained Android application behavior and permissions mappings was proposed. According to the classification of the characteristics of different applications,the concept minimal set of permissions of Android applications are given and the user recommend strategies was designed to help user take control of application permissions. Thereby forming a complete theoretical model based on Android application behavior permissions detection.Finally, based on the above analysis and research, the behavior detection system was designed and implemented. And by detecting the actual behavior of Android applications,the effect of the detection model was verified.