A Survey of Application of Artificial Intelligence Methods in SDN

With the rapid growth of data information, the pressure on network bandwidth is increasing day by day, and the traditional network architecture can no longer meet the daily needs of users. The birth and development of software-defined networking (SDN) brings new opportunities and challenges to the future network field. Due to its excellent flexibility and programmability, it has greatly improved the performance of the overall network. Coupled with the rapid development of artificial intelligence (AI) in recent years, researchers have begun to realize whether some methods of artificial intelligence can be integrated into SDN to further improve the utilization of network resources and overall network performance. This paper first expounds the background and progress of SDN and artificial intelligence; secondly, it summarizes the research of artificial intelligence in SDN from three aspects: intrusion detection, traffic classification and traffic scheduling; finally, it summarizes and looks forward to the future development trend of artificial intelligence and SDN.