An Enhanced Machine Learning based Data Privacy and Security Mitigation Technique: An Intelligent Federated Learning (FL) Model for Intrusion Detection and Classification System for Cyber-Physical Systems in Internet of Things (IoTs)

Authors

  • M. Aetsam Javed Department of Computer Science, Faculty of Computer Science & IT Superior University Lahore, 54000, Pakistan
  • Muhammad Ahmad National University of Sciences & Technology (NUST)
  • Junaid Ahmed Punjab Safe Cities Authority
  • Syed Muhammad Rizwan Punjab Safe Cities Authority, Department of Computer Engineering, University of Engineering and Technology Lahore, Pakistan
  • Anum Tariq University of Engineering and Technology Lahore

Abstract

The rapid development of industrial infrastructures with intelligent networking and computing technologies have dramatically increased the attack surface of cyber-physical systems (CPSs). Federated learning is a distributed learning method used to solve data security and privacy using machine learning, aiming to train global models together via multiple clients without sharing data. The rapid evolution of cyber threats poses significant challenges to modern cybersecurity systems and their associated legal frameworks. This paper addresses the problem of increasingly sophisticated breach methods that outpace traditional defense mechanisms. we propose an optimal federated learning Model, to detect and classify cyber threats against CPSs. Specifically for IoTs intrusion detection using federated learning framework. Data security and privacy received a great deal of research attention recently, as privacy protection becoming a key factor in the development of artificial intelligence based IOTs. It noted that security is a crucial issue in the End-to-End data security approach. The application of Machine Learning (ML) techniques to the well-known intrusion detection systems (IDS) is key to cope with increasingly sophisticated cybersecurity attacks through an effective and efficient detection process.  This paper provides a comprehensive exploration of Federated Learning and Virtual security techniques emphasizing their importance in modern networking for ensuring secure communication over untrusted networks like the internet. This study delves into various data security protocols such as PPTP, L2TP/IPsec, OpenVPN, IKEv2/IPsec, and WireGuard, evaluating their security mechanisms, strengths, and vulnerabilities. The paper also examines the emerging challenges in Machine Learning based model. In future a Deep learning based Fedrated  model can be used in detecting cyber threats to industrial CPSs by optimizing VPN usage to enhanced network security in an increasingly complex digital landscape.

Keywords: Cyber-physical systems (CPSs). Security Protocols, Encryption, OpenVPN, IKEv2/IPsec, WireGuard, Quantum Computing

Downloads

Published

2025-02-17

How to Cite

M. Aetsam Javed, Muhammad Ahmad, Junaid Ahmed, Syed Muhammad Rizwan, & Anum Tariq. (2025). An Enhanced Machine Learning based Data Privacy and Security Mitigation Technique: An Intelligent Federated Learning (FL) Model for Intrusion Detection and Classification System for Cyber-Physical Systems in Internet of Things (IoTs). Spectrum of Engineering Sciences, 3(2), 377–401. Retrieved from https://sesjournal.com/index.php/1/article/view/159

Issue

Vol. 3 No. 2 (2025): Spectrum of Engineering Sciences

Section

Articles