Enhancing Federated Learning Based IDS in Industrial Cybernetic physical Framework

Authors

  • Muhammad Arslan Faculty of Computing and Emerging Technologies, Emerson University Multan, Punjab, 60000, Pakistan
  • Wasif Akbar Faculty of Computing and Emerging Technologies, Emerson University Multan, Punjab, 60000, Pakistan

Keywords:

CNN, Cybersecurity, Deep-Learning, Industrial, CPS dataset

Abstract

The quick merging of old industrial systems with new networking and computer technologies (including 5G, software-defined networking, and artificial intelligence) has made industrial Cybernetic physical Framework a lot easier to hack. Still, it has been very hard to protect large, complex, and varied industrial Cybernetic physical Framework from cybersecurity concerns since there aren't many good examples of attacks. This research presents an innovative federated DL-Deep-Learning architecture, named , aimed at detecting cybersecurity concerns directed against industrial Cybernetic physical Framework. We first create a new DL-Deep-Learning-based IDS model for industrial Cybernetic physical Framework that uses a gated recurrent unit and a (CNN). Second, we set up a FL Framework that lets a lot of industrial Cybernetic physical Framework work together to construct a full IDS model while also protecting privacy. Comprehensive tests performed on an authentic industrial CPS dataset illustrate the significant efficacy of the proposed approach in identifying diverse cybersecurity concerns to industrial (CPS), as well as its superiority over existing leading techniques.

Downloads

Published

2025-09-01

How to Cite

Muhammad Arslan, & Wasif Akbar. (2025). Enhancing Federated Learning Based IDS in Industrial Cybernetic physical Framework. Machine Learning for Human Intelligence, 3(02), 8–18. Retrieved from https://mlhi.org/index.php/main/article/view/27

Similar Articles

1 2 > >> 

You may also start an advanced similarity search for this article.