Research And Realization On The Network Anomaly Flow Discovery Model-based Time Series

With the development of computer network technology and the expansion of network size, network-topology gets perplexing, network devices become complicated, and services-provided by the network are increasing. All these lead more faults and performance problems in the network. In order to detect network anomaly of a network in time, the alarms must be generated when the network is anomalous, for the manager to keep the network in right order. Therefore, networks must be monitored continuously. Network anomaly detection is a key part of network monitor, whether the network anomaly is detected accurately or not is very important to improve network availability and reliability.The prediction of the network traffic is an important part of the network performance management. An accurate prediction of network traffic can improve the effect of network management and network bandwidth utilization. This paper collected the campus network traffic by C/S mode,and the optimal model of the dynamic exponential smoothing model ,aiming the square of forecast errors, is established, through which the corresponding dynamic parameters can be gotten impersonally. It will enhance adaptability of exponential smoothing model on time sequence and solves the problems that smoothing parameter is static and to generate prediction deviation and so on.By testing, this model can more accurately predict the network traffic. So as to achieve the monitoring of campus network traffic and improve the quality of network services.This article thoroughly researched the present development of the network traffic anomaly detection technology, designed network traffic capture systems based on SNMP and Network Anomaly flow Discovery Algorithm based on times series. Research topics include the following aspects:(1)Design the system general framework, analyze managed objects in MIBs we need and design table structure which store these managed objects.(2)Collect data of the managed objects from SNMP agent using SNMP, and store these datas into database.(3) According to the characteristics that network traffic meet time series, Research network traffic anomaly discovery algorithm suitable for Inner Mongolia university of science and technology.(4)Show network running states by table or graph according to the analysis of collected data according to the collected datas.