An Abnormal Network Traffic Detection Scheme Using Local Outlier Factor in SDN
Nguyễn Ngọc Tuấn, Nguyen Danh Nghia, Pham Huy Hung, Dinh Khac Tuyen, Nguyen Minh Hieu, Nguyen Tai Hung, Nguyen Huu Thanh · 2021
Anomaly detection is an important step in the process of attack detection and mitigation. As the first step before attack detection mechanisms are used, it is to be decided if the network is in normal or abnormal state by anomaly traffic detection mechanisms. Software Define Network is a network architecture widely used recently as it is a flexible and open architecture allowing deploying on-demand functionalities into the network, including complex security and management algorithms. In this paper, we propose a machine learning algorithm based on Local outlier Factor (LoF) in Software Define Networking to detect abnormal traffic. LoF is a light-weight machine learning algorithm that requires minimal network resources. By setting up a testbed, we implement the proposed algorithm and evaluate its performance. The results show that the algorithm can perform in real-time and detect anomaly traffic in high accuracy.