Research On Anomaly Detection Of Network Traffic Based On Fractal Technology And Vector Quantization

With the rapid development of information technology, network-based applications and services are become more widespread. But the network intrusions and attacks have been become more pervasive and devastating, which threaten the security of Internet. It is so imperative that the robustness and security should been improved for network services and applications.In recent years, research on network anomaly detection based on fractal technology and vector quantization are two hot topics for network anomaly detection. In this paper, we firstly study the fractal–based algorithm for burst detection over data streams and the network anomaly detection method based on vector quantization. In order to achieve higher precision with less space and time complexity and to discover the abnormal network traffic and identify the cause of anomaly network traffic accurately and efficiently,this paper presents a new network traffic anomaly detection method:anomaly detection of network traffic based on fractal technology and vector quantization. It is shorted for the FB-VQ method.With this method we firstly preprocessed the network data in coarse-grained by FB algorithm for which have higher precision with less space and time complexity so that we can remain more useful features for more accurate network anomaly detection in next step. Then we detect the abnormal network traffic by using VQ method. The corresponding code word for each type of attacks are built with more detailed characteristics of data flow so that we can locate the reason of network anomaly accurately.Theoretical analysis and experiments show that the method can discover the abnormal network traffic and identify the cause of anomaly network traffic accurately and efficiently. With the certain innovation it can be served as a reference for the research of network anomaly detection.