Intrusion detection system in Software defined Network using machine learning approach - Survey

With the rapid development of internet in recent years, the networks are growing in size every day. As the networks are growing day by day, the task of managing the networks, effectively and efficiently also become a matter of concern. In case of traditional networks, due to their static nature, it is difficult to monitor the entire network efficiently and so now focus started shifting towards a new architecture that is Software defined networks. Being programmable in nature, SDN overcame a lot of shortcomings of traditional networks, however, the security of the network still remains the point of concern. The controller which is considered as the brain of SDN architecture can be taken as the prime target by attackers in order to attack the entire network. Therefore, in order to provide a robust mechanism for securing the networks, machine learning based intrusion detection systems came into picture. The proposed survey will provide an overview of software defined networks, intrusion detection systems and machine learning algorithms. The performance of machine learning algorithms that have been used recently by the researchers are compared. Also, this paper has highlighted the various challenges.