An Abnormal Traffic Discovery Way Based on EXBM Mechanism in Software Defined Network
Xi Wang, Yemeng Jiang · 2019
With the continuous development of the Internet new technologies, the importance of network security is becoming more and more important. In recent years, network traffic attacks on enterprise servers have been particularly serious. The emergence of SDN brings new solutions to the traffic detection of traditional networks, and domestic and foreign scholars combine the popular artificial intelligence algorithms to conduct research and modeling. Although the model based on machine learning can detect attacks to a certain extent, the applicability shows relatively poorly. This paper proposes a hybrid detection mechanism for DDoS attack: an abnormal traffic discovery way based on information entropy XGBoost model, called EXBM, and builds SDN environment to verify its effectiveness. The experimental results show that the detection accuracy of this scheme is improved compared with other algorithms.