Detection Of Distributed Denial Of Service Attacks Based On Network Traffic Characteristics Analysis

With the popularity and development of computer network,it increasingly become close to the people's daily lives. But network is a double-edged sword, it brings us convenience,but it also brings us a lot of trouble. Among the much issue, network security is priority. At present, network intrusion case happen frequently, various means of attack one after another.Denial of service(DOS) attacks has become one of the most common means of network attack,because it is simple and effective, its attack range is wider,and its destructive is large, subtle and defense hardly. It affects greatly efficient delivery service of the network and the host system.In particular, distributed denial of service(DDOS)attacks has brought a serious threat and immeasurable loss to the internet,because of its long incubation period, high Concurrent, more subtle, more destructive.But,nowadays seem to be no substantial improvement in anti-DDOS research on attack preventing,deteeting&retorting yet,nor did any efective method appear to predict the DDOS attacks in time.DDOS attacks detection and defending is one of the frontier in the field of network security.In this paper we propose a novel method(NTCAD) for detection of DDOS attacks by studying its architecture and the change of network traffic in the process of attack. And software model for the method is designed. In addition, Key issues of the choice of parameter and the cluster analysis of the sample is resolved. The simulation test result shows that the attack can be detected accurately and police error rate is very low. This method is simpler and easier to achieve than the traditional detection method.