Genetic Algorithm based Intrusion Detection System for Software-Defined Network Architecture
Xuejian Zhao, Songle Chen, Yunfeng Yu, Zhixin Sun · 2020
Intrusion detection systems (IDSs) are widely used to protect communication networks against traffic attacks. However, traditional IDSs may not be effective in a software-defined network that has a centralized controller to manage massive amount of data traffic. In this paper, we propose a novel IDS model to meet the requirements of software-defined network (SDN). In particular, our proposed system collects and analyzes traffic mainly at the control plane of SDN. However, once the amount of data transferred exceeds the data processing capacity of the IDS, network congestion will occur. We tackle this issue with a probability based traffic sampling algorithm. The optimal solution, i.e., the sampling probability of each sampling point, is approached with the genetic algorithm (GA). The simulation results demonstrate that our proposed GA-based IDS model can improve the detection efficiency in SDN.