Overview of DDoS Attack Detection in Software-Defined Networks

Abstract

Software Defined network (SDN), as a new network architecture, is the direction of future network development. By decoupling the control plane and data plane of the network, the control and management of network resources can be improved. However, SDN centralized controllers become the target of malicious attacks, prone to the risk of single point of failure, of which DDoS attacks are the main attack behavior. Through extensive literature investigation, this paper systematically reviews the latest progress of DDoS attack detection in SDN environment. Firstly, the SDN architecture and corresponding DDoS attacks are described, and the commonly used data sets and evaluation indicators are summarized. Secondly, the data preprocessing technology is summarized, and the data dimensionality reduction technology is introduced in detail. Then, at the level of detection technology, it focuses on the advantages and disadvantages of detection algorithms based on statistical analysis, machine learning and deep learning. Finally, the challenges of current research are analyzed, and the future development direction is forecasted. In order to provide reference for future research and development, at the same time, it is of great significance to improve the safety of SDN products in the future.

Keywords

• SDN
• DDoS
• entropy
• machine learning
• deep learning