ARTIFICIAL INTELLIGENCE-BASED INTRUSION DETECTION FOR DDOS ATTACKS IN SOFTWARE-DEFINED NETWORKING
Software Defined Networking (SDN) is a software-oriented network concept created by the separation of control and data planes. Furthermore, SDN is vulnerable to a variety of security threats, including distributed denial of service (DDoS) attacks. This article presents a flexible modular software architecture that enables the identification and mitigation of DDoS attacks. Normal traffic has been observed. Subsequently, a DDoS attack was executed and observed. This article also introduces a system that detects and prevents DDoS attacks in SDN using Machine Learning techniques. Numerous test cases were executed with varying scenarios using Mininet. The Decision Tree model has a 91 percent accuracy rate. The Decision Tree model outperformed others such as Artificial Neural Networks (ANN), which had a 78% accuracy rate, and Naive Bayes, with 68%. Consequently, there was a consensus that the most appropriate model for DDoS detection was the Decision Tree. Artificial intelligence models used in the system can be easily reintegrated based on the success rate of these models trained on Google Colaboratory. This feature is aimed at saving a significant amount of time during the training process. Real-time DDoS attacks have been detected. After the detection process, the computer/system affected by the attack was also secured by the controller to prevent further impact.
Сведения
- Опубликовано
- 2026 · Т. 357, № 1
- Язык
- Английский
- ISSN
- 1991-346X · 2518-1726
- Издатель
- Central Asian Academic Research Center LLP