DDoS Attack Detection System Based on GBDT Under SDN

DDoS attacks is one of the major threats to network security today. The emerging network architecture SDN has the characteristics of centralized control and programmability, which brings convenience and flexibility to the network management, but these characteristics make it more vulnerable to malicious attacks, resulting in network paralysis. To address this problem, this paper proposes a GBDT-based detection method to detect the attack traffic in the network by combining the machine learning approach. The experimental results show that the GBDT achieves higher detection efficiency compared to other methods, with an average detection accuracy of 93.88% and a false alarm rate of 0.0411.