Research And Implementation Of Android-based Mobile Communications Network Intrusion Detection System

With the development and application of3G networks, the service and contentsprovided by mobile communications system increase dramatically. While it gives usersan unprecedented experience, it also proposes more challenges to the security of themobile communication network.In order to improve the safety performance of the mobile communication networkterminal, this paper designs and implements an Android side intrusion detection systemby studying the security of the currently dominant Android mobile operating system.The system uses the C/S mode what can effectively relieve the constraints of thehardware resources of the mobile terminals and rapid detection mechanism what candetect known malware accurately and efficiently. For unknown software, we first userole-based static detection method, if the detection result is insecure software, we willuse behavior-based dynamic detection to further affirm the security of this software. Thesystem has higher detection efficiency and detection accuracy through thelayer-by-layer detection.The same time, this paper collects a large number of samples of malicious softwareand general application, and analyzes the efficiency, accuracy and functionality of thesystem through the comparison with other security software. As can be seen from thecomparison results, rapid detection system can quickly detect known malware, theresponse time is ideal; the detection accuracy of static detection system in more than73%on average, compared to the accuracy of other security software is higher; dynamicdetection system is more comprehensive in terms of detection function.In all security threats faced by the mobile communication network, the securitythreat of wireless interface is the most important in addition to the terminal threat. Thispaper expounds that3G network can effectively prevent unauthorized access fromunauthorized users and make it impossible to carry out an attack by analyzing the3Gnetwork access security mechanisms, but3G network could not prevent internal attacksfrom legitimate users. In light of this, this paper presents an ANN-based intrusion detection model,designs and implements its key module. The model acquires communication behaviorprofile of users by learning and training the user’s communication behavior. Then it candetect abnormal behaviors of legitimate users and prevent network internal attacks fromlegitimate users. It further enhances the security of the mobile communication network.