Software-Defined-Networking-Enabled Traffic Anomaly Detection and Mitigation

Traffic anomaly detection has been a principal direction in the network security field, which aims to identify attacks based on significant deviations from the established normal usage profiles. Recently, a new networking paradigm, software defined networking (SDN), has emerged to facilitate effective network control and management. In this paper, we present the advantages of leveraging SDN to detect traffic anomaly, and review recent progresses in this direction. Despite their effectiveness for traditional traffic, SDN-based traffic anomaly detection methods have to face the challenge of continuously increasing network traffic. To this end, we propose two refined algorithms to be used in an anomaly detection framework which can handle voluminous data, and report some experimental results to demonstrate their performance.