Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Sep 4;20(9):e0331470.
doi: 10.1371/journal.pone.0331470. eCollection 2025.

Multilayered SDN security with MAC authentication and GAN-based intrusion detection

Nanavath Kiran Singh Nayak  1 , Budhaditya Bhattacharyya  1
Affiliations

Multilayered SDN security with MAC authentication and GAN-based intrusion detection

Nanavath Kiran Singh Nayak et al. PLoS One. .

Abstract

Computer networks are highly vulnerable to cybersecurity intrusions. Likewise, software-defined networks (SDN), which enable 5G users to broadcast sensitive data, have become a primary target for vulnerability. To protect the network security against attacks, various security protocols, including authorization, the authentication process, and intrusion detection techniques, are essential. However, there are several intrusion detection strategies, but the most prevalent methods show low accuracy and high false positives. To overcome these problems, this research work presents a novel four-Q curve authentication system based on Media Access Control (MAC) addresses for a multilayered SDN intrusion detection system utilizing deep learning techniques to identify and prevent attacks. The Four-Q curve authentication system leverages elliptic curve cryptography, a high-performance algorithm that improves authentication security and computational efficiency. Initially, Four-Q curve authentication is performed, followed by univariate ensemble feature selection to select optimal switches. Then, the data collected through the switches are classified as normal, assault, and suspect packets based on the Dual Discriminator Conditional Generative Adversarial Network (DDcGAN) approach. Further, the optimization of DDcGAN is accomplished using the Sheep Flock Optimization Algorithm (SFOA), whereas suspicious packets are categorized using the Growing Self-Organizing Map (GSOM). The DDcGAN-based intrusion detection system outperforms the state-of-the-art approaches in terms of accuracy, precision, F1 score, sensitivity, false-positive rate, power consumption, and network throughput. It achieved an accuracy of 98.29%, an F1 score of 0.975, and a precision of 95.8%. The system's true positive rate attained 99.04% at 50% malicious nodes, while the false alarm rate was as low as 2.05% under the same conditions. Moreover, the system exhibits 4.5% energy savings when compared to existing approaches.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Architecture of proposed system model.
Fig 2
Fig 2. System model of the proposed method.
Fig 3
Fig 3. Schematic representation of univariate ensemble-based feature selection.
Fig 4
Fig 4. Flow diagram for the proposed methodology.
Fig 5
Fig 5. Network simulation of the proposed method.
Fig 6
Fig 6. Accuracy analysis of DDcGAN-GSOM with existing techniques.
Fig 7
Fig 7. F1-score analysis of DDcGAN-GSOM with existing methods.
Fig 8
Fig 8. Precision analysis of DDcGAN-GSOM with current approaches.
Fig 9
Fig 9. Analysis of DDcGAN-GSOM delay with existing techniques.
Fig 10
Fig 10. Energy consumption analysis of DDcGAN-GSOM with existing methods.
Fig 11
Fig 11. Energy consumption analysis of DDcGAN-GSOM with existing approaches.
Fig 12
Fig 12. Throughput analysis of DDcGAN-GSOM with prevailing approaches.
Fig 13
Fig 13. True positive rate (TPR) analysis of DDcGAN-GSOM with existing techniques.
Fig 14
Fig 14. False alarm rate analysis of DDcGAN-GSOM with existing approaches.
Fig 15
Fig 15. Specificity comparison of DDcGAN-GSOM with prevailing methods.
Fig 16
Fig 16. Discriminator loss of suggested technique.
Fig 17
Fig 17. Generator loss of the proposed method.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Nayak NKS, Bhattacharyya B. An Intrusion Detection System for 5G SDN Network Utilizing Binarized Deep Spiking Capsule Fire Hawk Neural Networks and Blockchain Technology. Future Internet. 2024;16(10):359. doi: 10.3390/fi16100359 - DOI
    1. Ravi V, Chaganti R, Alazab M. Recurrent deep learning-based feature fusion ensemble meta-classifier approach for intelligent network intrusion detection system. Computers and Electrical Engineering. 2022;102:108156. doi: 10.1016/j.compeleceng.2022.108156 - DOI
    1. Yungaicela-Naula NM, Vargas-Rosales C, Pérez-Díaz JA, Carrera DF. A flexible SDN-based framework for slow-rate DDoS attack mitigation by using deep reinforcement learning. Journal of Network and Computer Applications. 2022;205:103444. doi: 10.1016/j.jnca.2022.103444 - DOI
    1. Naqash T, Shah SH, Islam MNU. Statistical Analysis Based Intrusion Detection System for Ultra-High-Speed Software Defined Network. Int J Parallel Prog. 2021;50(1):89–114. doi: 10.1007/s10766-021-00715-0 - DOI
    1. Birkinshaw C, Rouka E, Vassilakis VG. Implementing an intrusion detection and prevention system using software-defined networking: Defending against port-scanning and denial-of-service attacks. Journal of Network and Computer Applications. 2019;136:71–85. doi: 10.1016/j.jnca.2019.03.005 - DOI

LinkOut - more resources