Using Machine Learning in Software Defined Networks to Recognize and Avoid DDOS Attacks

Abstract

Abstract: In recent years, Distributed Denial of Service (DDoS) attacks have become a significant concern for network administrators and internet service providers. DDoS attacks are designed to overwhelm a target network with a flood of traffic, causing it to crash or become unavailable. These attacks can have severe consequences, including financial losses, reputational damage, and disruption of critical services. In response to this growing threat, researchers have been exploring various techniques to detect and mitigate DDoS attacks. One promising approach is to use machine learning (ML) algorithms in software-defined networks (SDN). SDN provides a centralized control plane that enables fine-grained control over network traffic, making it an ideal platform for deploying ML-based DDoS detection and mitigation techniques. This paper discusses the use of ML in SDN to recognize and avoid DDoS attacks, providing an overview of the key concepts, techniques, and challenges involved in this emerging field. Six ML algorithms were assessed, viz. Logistic Regression (LR), Naive Bayes (NB), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The inspection showed that KNN, DT and RF is the best appearing ML algorithm. Results showed that our model is able to recognize attacks precisely and rapidly.