Machine learning in SDN networks for secure industrial cyber physical systems: a case of detecting link flooding attack
by Priyanshi Deliwala; Rutvij H. Jhaveri; Sagar Ramani
International Journal of Engineering Systems Modelling and Simulation (IJESMS), Vol. 13, No. 1, 2022

Abstract: Software-defined networking (SDN) is an emerging network architecture which has potential to solve real-time challenges of industrial cyber physical systems (ICPSs). However, it also opens a gateway to different network attacks such as link flooding attack (LFA). This paper illustrates how SDN's control layer is endangered to LFA. While the proposed approach (Iwendi et al., 2020a) can appear suitable on the surface, some weaknesses and anomalies are discovered when deliberated deeper. In the current paper, we point out these anomalies and the limitations of those anomalies by applying two machine learning algorithms, namely ANN-MLP and random forest to correctly classify the virulent traffic during congestion. We carry out experiments on Mininet network emulator and use the WEKA tool to assess the metrics by utilising the available datasets. The results show that a random forest can accurately detect virulent traffic and efficiently terminate it under congestion.

Online publication date: Mon, 09-May-2022

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Engineering Systems Modelling and Simulation (IJESMS):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com