A DDoS Attack Detection Model Based on Machine Learning Algorithm in SDN Environment

The Software Defined Network(SDN)is an emerging network architecture that separates control logic from forwarding logic.In SDN, the controller has a global control of the network.Because of this feature of the controller, making it becomes the main goal of the distributed denial of service(DDoS)attack.Aiming at this problem, this paper proposes a method based on machine learning to detect the DDoS attack model.Firstly, it uses the entropy to check whether the traffic is abnormal.After extracting the abnormal alarm, the network flow feature is extracted, and SVM and K-means are called to detect the DDoS attacks.The experimental results show that the proposed algorithm can reduce the false alarm rate, and the detection rate and accuracy of DDoS attacks are higher than those of the original SVM and K-means.In addition, the experimental results show that the average CPU utilization rate of the proposed model is lower than that of SVM + K-means without entropy detection