A Comprehensive Review of SDN Intrusion Detection Using Ensemble Machine Learning

Abstract

Software Defined Networking (SDN) has developed as a potential networking paradigm that improves network programmability, flexibility, and centralized management. The centralized architecture of SDN presents considerable security challenges, rendering Intrusion Detection Systems (IDSs) crucial for detecting and addressing malicious actions. Conventional IDS methodologies frequently experience elevated false alarm rates and exhibit restricted efficacy against advanced and emerging cyber threats. To tackle these issues, machine learning approaches, especially ensemble learning methods, have garnered significant interest for their capacity to enhance detection performance and generalization ability. This article offers a thorough examination of methodologies for intrusion detection using machine learning in software-defined networking contexts, emphasizing ensemble learning strategies. The study investigates prevalent datasets, assesses the advantage and drawbacks of current methodologies, and analyzes contem