Protecting User’s Privacy By Analyzing The Combination Of Application’s Permissions On Android System

Protecting user’s privacy is always the research focus in smartphone security, especially as Android system is getting widely used and usually most users are not pay attention to the application’s permissions at install time, the leakage of user’s privacy becomes much worse. Therefore, we need a solution to check and analysis the applications which are being installed, to assist users find applications which may leak their privacy.For protecting user’s privacy, there have been some works, including investigation and analysis of smartphone users’use habits, analyzing applications’ configuration files, modifying and extending the PackageInstaller on Android system, static analysis and dynamic analysis and so on. However, most of these works are proposed as professional application analysis tools, they do not take the user’s usability into consideration, and some solutions which modifying and extending the PackageInstaller have the problems of low efficiency and user understandability. So in this paper, we propose a solution to protect user’s privacy by analyzing the combination of application’s permissions. It realizes this purpose by analyzing the application’s request permissions set at install time.This user privacy protection solution takes the efficiency of analyzing the application’s permissions, user’s understandability and the efficiency of protecting user’s privacy into consideration, and basing on these we implement the solution to protect user’s privacy. The major works and innovation of this paper are as follows:(1) We can find the permission combinations which may leak user’s privacy by analyzing all the permissions on the latest Android system.(2) Through modifying and extending the PackageInstaller on Android system, we can add an analyzing module, which can analyze the application’s permissions at install time. It analyses the application’s request permissions and find the permission combinations that may leak user’s privacy, then it displays the analysis result to the user in an easy-understanding way, which will help the user make a decision about whether or not to install the application.(3) We give every permission combination a certain weight, so the analysis result will be red as a score and a level, which imply the possibility of leaking the user’s privacy by the application. By adopting this method, the user can have a better cognition of the result.(4) By modifying some system methods in Android source codes, we provide users the ability to dispose certain permission combinations that may leak their privacy, and then they can protect their privacy.