摘要
针对Android应用中普遍存在的隐私窃取问题,提出一种基于动态信息流的Android恶意应用检测方法。定制合适的Android内核,在应用运行过程中收集相关的文件操作与进程通信信息;基于隐私数据集构建动态信息流模型;通过跟踪动态信息流,判定隐私数据是否被发送、是否导致信息泄露。大规模的实验结果表明,该方法对隐私窃取类应用检出率高达94.57%,验证了其有效性。
The increasing presence of Android privacy leakages poses a significant private risk for Android smartphone users.An Android malware detection method based on dynamic information flow was proposed.A fit Android kernel was made,which collected related operations of files and the information of process communication.Dynamic information flow model based on private data was constructed.By tracking the dynamic information flow,whether the private data was sent and whether it caused information leakage were judged.Results of large scale experiments show that the detection rate of the proposed method reaches 94.57% for privacy leakage applications.The results verify the good detection capacity of the proposed method.
出处
《计算机工程与设计》
北大核心
2017年第10期2646-2651,共6页
Computer Engineering and Design