DroidBet:事件驱动的Android应用网络行为的自动检测系统

发布时间：2018-08-06 18:07

【摘要】：多数Android应用需要通过连入互联网与外界进行通信,所有与网络相关的活动都涉及网络流量,通过分析建模Android应用的网络流量,可以一定程度上掌握Android应用的网络行为。因此,设计了一个事件驱动的网络行为自动检测系统DroidBet,来对Android应用进行自动测试评估。首先,建立一个场景模拟事件库,用来模拟应用程序运行过程中可能执行的事件,从而尽可能地触发应用程序的网络行为;然后,自动生成基于状态转移分析方法的测试序列,同时对应用程序测试过程中的网络行为进行动态收集;最后,采用机器学习方法对收集到的网络行为进行学习、训练,生成基于BP神经网络的网络行为模型,实现对未知的Android应用的行为检测。实验结果表明,DroidBet能够有效地触发并提取应用程序的网络行为,并具有准确度高、系统资源开销低等优点。
[Abstract]:Most Android applications need to communicate with the outside world by connecting to the Internet. All network related activities involve network traffic. By analyzing the network traffic of modeling Android applications, the network behavior of Android applications can be mastered to a certain extent. Therefore, an event driven automatic detection system of network behavior, DroidBet, is designed. An automatic test evaluation for Android applications. First, a scene simulation event library is built to simulate events that may be executed during the application process, thus triggering the network behavior of the application as much as possible; then, the test sequence based on the state transfer analysis method is automatically generated and the application test process is also tested. The network behavior is collected dynamically. Finally, the network behavior collected by the machine learning method is learned and trained, and the network behavior model based on BP neural network is generated to realize the behavior detection of the unknown Android application. The experimental results show that DroidBet can effectively trigger and extract the network behavior of the application, and have the network behavior of the application. It has the advantages of high accuracy and low system resource overhead.
【作者单位】：南京理工大学计算机科学与工程学院;
【基金】：国家自然科学基金资助项目(No.61472189)~~
【分类号】：TP393.06

【相似文献】