Research On IOS Application Vulnerability And Malicious Behavior Detection

With the development of mobile internet,mobile intelligence devices have played a pivotal role in people's daily work and life.More and more users use mobile intelligence devices to storage personal privacy information.In order to keep the user's private information safe,the application must secure itself at fir-st.Since there are only a few contents for the application's security detection,they cannot provide a comprehensive security assessment for the application.Secondly,the application market has to guarantee the detection rate of malicious applications in the process of application review.The defects for existing iOS malicious application detection algorithms are at low speed and accuracy,which bring challenges to accuracy and efficiency for a larger number of iOS malicious application detection.According to the limitations of existing work on iOS application security,this thesis is divided into two parts,which are the detection of iOS application vulnerability and iOS malicious application detection.The contribution of this thesis as follows:First of all,this thesis proposes multinomial detection content for iOS application static security and implements automatic detection.By analyzing the security risks of iOS application,combined with the characteristics of the iOS application,this thesis put forward a detection mechanism for the risks and implements automatic detection,which improved the existing content of static detection for iOS application security.Secondly,this thesis put forward iOS application dynamic security detection content,designing the detection scheme for dynamic detection content.By analyzing security risks of iOS application during runtime,this thesis proposing the content of dynamic security detection and developed a detection scheme.After the dynamic and static security detection,a security evaluation scoring mechanism is proposed to score the application.Last but not least,this thesis proposes to take the permission usage of the iOS application as one of the features of iOS malicious application detection.Irn comparative experiments,this paper demonstrated that the newly added feature can improve the detection accuracy of the static detection module.This thesis also proposed a combined dynamic and static iOS malicious application detection algorithm.The static detection module was used for the preliminary screening of the applications under test,which greatly reduced the number of applications under test in the dynamic detection module,in the result,speeding high the detection of iOS malicious applications.