Design And Implementation Of User Abnormal Behavior Analysis System In Identifier Private Network

With the rapid development of network technology, it brings a lot of convenience to people’s life, but also brings more and more security problems. Some new network architecture, which is represented by the Universal Identifier Network, overcomes the original defects of the traditional network, and has better security and scalability properties. This thesis is based on the Major Security Project "Applied Research of Identifier Network Technology in Mobile Private Network". We designed and implemented a user abnormal behavior analysis system in the Identifier Private Network which is an integrated network of Universal Identifier Network and GPRS network. It improved users’ controllability and manageability, and further ensured the security of the network.This thesis is mainly about design and realization method of user abnormal behavior analysis system in Identifier Private Network. This system can not only detect common anomalies in Identifier Private Network, but also locate the abnormal users. It achieved flexible controllability of user behavior through reputation and level adjustment. According to the demand of anomaly detection in Identifier Private Network, the paper adopted NetFlow capture technology and entropy-based traffic quantization method to realize data collection and pre-processing. And we used a SVM multi-class classification method to detect anomaly traffic. According to the demand of user management of Identifier Private Network, we put forward a user reputation management mechanism. It provided extra protection for the Identifier Private Network upon the already-existing protection mechanism.First, this thesis gave an overview of Identifier Private Network, and introduced the integration method of Universal Identifier Network and GPRS network, and introduced user’s access authentication process in the Identifier Private Network. Second, we analyzed the detailed requirements of the user abnormal behavior analysis system and put forward an overall design scheme and block diagram, and then divided the system into some main function modules including data collection and output module, data analysis and processing module, and user reputation management module. After that, we gave detailed description of each function module and its sub modules, and explained the realization of each function module from the perspective of code.Finally, the actual functions of main modules of the system were tested. The results showed that the implemented system can detect and discover users’ common abnormal behavior in Identifier Private Network, and can manage the reputation and level of users through the user reputation management mechanism, and it achieved flexible control of user behavior in Identifier Private Network.