Intrusion Detection in Software-Defined Networking Using Machine Learning Models

Software-defined networking (SDN) is a new networking paradigm developed to reduce network complexity via control and management of the network from a centralized location. Nevertheless, the dynamic nature of SDN can lead to many vulnerabilities and threats, including denial of service (DOS) and Distributed Denial of service (DDoS) attacks. Thus, deploying Intrusion Detection Systems (IDSs) based on machine learning (ML) is a crucial part of the network architecture to monitor malevolent activities. This paper compares three ML models for intelligent intrusion detection in SDN: support vector machines, K-Nearest Neighbors, and Naive Bayes Networks. To evaluate and measure the performance of ML models, we used the DDoS-SDN dataset and compared their evaluation metrics, such as accuracy.