Edge DDoS Attack Detection Method Based on Software Defined Networks

Edge computing extends the traditional cloud computing architecture by using the computing and storage resources on the edge of the network, making people's work and life more convenient. However, these devices at the edge of the network are widely distributed and the environment is relatively complex. Attackers use these vulnerable IoT devices to build botnets to initiate distributed denial of service attacks, posing a serious threat to the normal use of such networks. In response to this problem, we propose an anomaly detection framework based on software-defined networking (SDN). The edge controller in the SDN network is used to obtain the flow information and extract the features of the flow. The XGBoost algorithm optimized by genetic algorithm (GA-XGBoost) we proposed is used to classify and detect the flow. Experimental results show that compared with other algorithms of the same type, our proposed algorithm has a lower false alarm rate and higher accuracy.