LDoS Attack Detection System Based on ACO-LightGBM

Software Defined Network (SDN) is a new architecture that separates the data layer from the control layer.However, this architecture of SDN is not enough to resist all denial of service attacks. For example, low-rate denial of service attacks (LDoS) are periodic, low-rate, and very hidden. To solve this problem, this paper designs a real-time attack detection system on the SDN architecture. LightGBM optimized by Ant Colony Optimization (ACO-LightGBM) is a primary tool for LDoS attack detection, and Mininet is used to simulate the real LDoS attack environment. Experiments show that LightGBM optimized by ant colony algorithm has the best detection performance compared with several other classifiers.