Version 1
: Received: 15 August 2023 / Approved: 15 August 2023 / Online: 15 August 2023 (09:40:34 CEST)
How to cite:
Tavangari, S.; Taghavi Kulfati, S. Review of Advancing Anomaly Detection in SDN through Deep Learning Algorithms. Preprints2023, 2023081089. https://doi.org/10.20944/preprints202308.1089.v1
Tavangari, S.; Taghavi Kulfati, S. Review of Advancing Anomaly Detection in SDN through Deep Learning Algorithms. Preprints 2023, 2023081089. https://doi.org/10.20944/preprints202308.1089.v1
Tavangari, S.; Taghavi Kulfati, S. Review of Advancing Anomaly Detection in SDN through Deep Learning Algorithms. Preprints2023, 2023081089. https://doi.org/10.20944/preprints202308.1089.v1
APA Style
Tavangari, S., & Taghavi Kulfati, S. (2023). Review of Advancing Anomaly Detection in SDN through Deep Learning Algorithms. Preprints. https://doi.org/10.20944/preprints202308.1089.v1
Chicago/Turabian Style
Tavangari, S. and Somayeh Taghavi Kulfati. 2023 "Review of Advancing Anomaly Detection in SDN through Deep Learning Algorithms" Preprints. https://doi.org/10.20944/preprints202308.1089.v1
Abstract
Recent SDN advances address traditional network management challenges through centralized control and plane separation. SDN prevents breaches using a centralized controller but introduces risks. The controller can be a single point of failure. Thus, an OpenFlow Controller's flow-based anomaly detection enhances SDN security. Our research explored two OpenFlow intrusion detection methods. The first employed machine learning, NSL-KDD dataset, and feature selection, yielding 82% accuracy with random forest. The second combined deep neural networks with GRU-LSTM, achieving 88% accuracy using ANOVA F-Test and feature elimination. Experiments highlighted deep learning as superior for OpenFlow intrusion detection.
Keywords
SDN; machine learning; algorithms; GRU-LSTM
Subject
Computer Science and Mathematics, Computer Science
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.