The Research And Implementation Of System For Network Anomaly Detection

With the faster development of the network technology, various kinds of new crises have appeared. But the people's request for the network service is improving constantly. It has become a urgent problem how to carry on effective network management.The article analyses the limitation of network management products and the research on anomaly situation of network, proposes the network anomaly detection system. The system is designed for the large LAN such as company network, campus network. It can monitor the equipment of different manufactures and catch the anomaly situation of the network in time, help network manager understanding the performance of network better, offer more information for network manager to diagnoses and get rid of the trouble.The article analyses numerous network parameters and finds out the parameters having closed contacts with the network behavior, and then establishes the rules of judging anomaly situation according to the parameters. The article discusses of the network flow model and proposes the linear regression model.It researches the dependence among the alert information in addition.The network anomaly detection system is made of three modules mainly: data collection module, anomaly judgement module and alert analyse module. Data collection module collectes the data of the selected parameters and save; anomaly judgement module analyses the data according to the rule, if found any anomaly situation, the module will make alerts; alert analyse module analyses the dependce of these alerts.