LAUSR

In recent years, there are more and more abnormal activities in the network, which greatly threaten network security. Hence, it is of great importance to collect the data which indicate the running statement of the network, and distinguish the anomaly phenomena of the network in time. In this paper, we propose a novel anomaly network traffic detection algorithm under the cloud computing environment. Firstly, the framework of the anomaly network traffic detection system is illustrated, and six type of network traffic features are consider in this work, that is, (1) number of source IP address, (2) number of source port number, (3) number of destination IP address, (4) number of destination port number, (5) Number of packet type, and (6) number of network packets. Secondly, we propose a novel hybrid information entropy and SVM model to tackle the proposed problem by normalizing values of network features and exploiting SVM detect anomaly network behaviors. Finally, experimental results demonstrate that the proposed algorithm can detect anomaly network traffic with high accuracy and it can also be used in the large scale dataset.