Electronic Thesis/Dissertation
 

Enhancing Cyber Resiliency of Industrial IoT Attack Detection

Open Access Deposited

Industrial Internet of Things (IIoT) devices have become an essential component of critical infrastructure environments to improve efficiency and productivity. Traditional intrusion detection systems are not designed for the unique security risks of IIoT devices and the expanding threat landscape. Machine learning-based intrusion detection systems have evolved to address this challenge but require access to sensitive network traffic data, compromising confidentiality and privacy. Privacy-preserving machine learning models are proposed in this praxis to enhance the resiliency of attack detection for Industrial Internet of Things network traffic. Fully homomorphic encryption is implemented as the privacy-preserving technique for the attack detection models to enable inference on encrypted IIoT network traffic data. This praxis utilizes the publicly available Edge-IIoTset dataset to implement and assess encrypted inference with three machine-learning models: Multilayer Perceptron Artificial Neural Network, eXtreme Gradient Boost, and Random Forest. The results of this praxis show that privacy-preserving machine learning models can detect attacks in encrypted industrial IoT network traffic and contribute to enhancing resiliency in detection without compromising confidentiality.

Author Language Keyword Date created Type of Work License
  • All rights reserved
Rights statement GW Unit Degree Advisor Committee Member(s) Persistent URL

Notice to Authors

If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.

Thumbnail Title Date Uploaded Visibility Actions
Preview of Cameron_gwu_0075A_17089.pdf Cameron_gwu_0075A_17089.pdf 2025-04-09 Open Access