ML & DL - based DDoS Detection in SDN: Implementation and Comparative Analysis
Author(s): Aarti Jadav, Parul Sharma
Publication #: 2604003
Date of Publication: 25.04.2026
Country: India
Pages: 1-11
Published In: Volume 12 Issue 2 April-2026
DOI: https://doi.org/10.62970/IJIRCT.v12.i2.2604003
Abstract
Software Defined Networking (SDN) is a modern networking approach that separates the control plane from the data plane, allowing centralized control and better network management. However, the centralized SDN controller is highly vulnerable to Distributed Denial of Service (DDoS) attacks, where attackers send a large volume of malicious traffic to overload the controller and disrupt network services. Traditional DDoS detection methods are not effective in dynamic network environments because they rely on predefined rules and signatures, which makes it difficult to detect new and unknown attacks.
To overcome this limitation, Machine Learning (ML) and Deep Learning (DL) techniques are used to detect DDoS attacks by analyzing network traffic patterns. In this research, various ML algorithms such as Random Forest, Decision Tree, Support Vector Machine, and K-Nearest Neighbor, as well as DL models such as Convolutional Neural Network, Long Short-Term Memory, and Multi-Layer Perceptron, are implemented and compared for DDoS detection in SDN. The performance of these models is evaluated using metrics such as accuracy, precision, recall, F1-score, training time, and prediction time to identify the most efficient model for real-time DDoS detection.
Keywords: Keywords: Software Defined Networking (SDN), DDoS Detection, Machine Learning Algorithms, Deep Learning Techniques, Network Security, Intrusion Detection System, Network Traffic Analysis.
Download/View Count: 4
Share this Article