International journal of engineering science and management (IJESM)
Enhancing SDN Security: Machine Learning-based Detection of Network Intrusion Attacks
| © 2024 by IJESM Journal |
| Volume-24 Issue-1 October |
| Year of Publication : 2024 |
| Author :Indrabayu, Taslinda, Rizka Irianty, Sitti Wetenriajeng Sidehabi |
| DOI : XX.XXXXX/XXXXXXXX/IJESM-XXXXXXXXX |
Abstract
The rapid proliferation of Software-Defined Networking (SDN) has introduced numerous advantages for managing complex networks. However, this paradigm shift has also escalated the risks of network intrusion attacks, demanding advanced security measures to safeguard critical infrastructures. Traditional security approaches often fall short in addressing the dynamic and sophisticated nature of modern cyber threats, necessitating the integration of intelligent systems for enhanced detection and mitigation capabilities. This paper investigates the application of Machine Learning (ML) techniques to bolster the security framework of SDN against various forms of network intrusion attacks. By leveraging the inherent capabilities of ML algorithms, such as deep learning and anomaly detection, the proposed model aims to fortify the SDN architecture by continuously monitoring network traffic patterns, identifying anomalies, and predicting potential intrusion events with greater accuracy and efficiency. Through a comprehensive analysis of real-time network data and simulations, this research demonstrates the efficacy of the ML-based approach in detecting both known and unknown intrusion attempts, thereby mitigating potential threats before they can compromise the network’s integrity. Furthermore, the integration of ML-driven security measures fosters adaptability to evolving attack strategies and enables proactive responses to emerging threats, ensuring a robust and resilient SDN infrastructure. The findings underscore the significance of integrating intelligent security mechanisms within SDN environments to proactively detect, prevent, and mitigate intrusion attacks. By harnessing the power of Machine Learning, this study contributes to the development of a more secure and reliable SDN ecosystem, fostering trust and confidence in the integrity of modern network infrastructures. A DDoS attack occurs when multiple systems collaborate to target a specific host simultaneously. In SDN, the control layer software, situated between the application and infrastructure layers, governs the devices within the infrastructure layer. To identify and thwart malicious traffic, we propose a machine learning-based approach utilizing Random Forest, AdaBoost, CatBoost, and XGBoost algorithms in this study. Our experimental results demonstrate that the Random Forest, CatBoost, and XGBoost algorithms exhibit superior detection rates and accuracy, showcasing their efficacy in fortifying SDN against potential security breaches.