A Review of Intrusion Detection Systems in Software Defined Networks
Athira Vijayan, A. Anitha · 2025
Software-defined networking (SDN) is a novel networking technique that divides controllers from network devices, like switches as well as routers. The centralized model of SDN makes complete network management and identifies the needs of present data centers. However, SDN model presents maximum advantages; new attack risk is a crucial issue that can stop extensive usage of SDNs. The SDN controller is a vital factor; it is a striking aim for intruders. Traffic can be routed by SDN controller, on the basis of individual necessities that cause rigorous damage to the complete network in the scenario of an attacker effectively accessing SDN controller. Moreover, in the SDN, intrusion detection requires more and more ML approaches, which need to be discovered and consider criticality of SDN controller. In the SDN, the experimentation exhibits that IDS is a hot topic and various approaches presented and are subjugated by Machine Learning (ML) that employs the traffic flow of network to recognize abnormal behaviours. Thus, this work performed a survey of a few IDS models in an SDN environment. In SDN, the IDSs are developed with an ML model; nonetheless, a Deep Learning model is being explored to attain better accuracy and efficiency. Moreover, brief research is conducted on a few benchmark databases employed to design IDS in SDN model. Finally, a discussion is provided that sheds light on incessant issues and IDS challenges for security of SDN to conclude the review.