Modern buildings, transport systems, and enterprises increasingly rely on Internet-of-Things (IoT) and Operational Technology (OT) devices such as cameras, smart sensors, lighting systems, and alarms. These devices are often produced by different manufacturers and deployed at large scale, making them difficult to monitor and secure. As a result, cyber-attacks targeting these devices can disrupt essential services and critical infrastructure.
Current approaches for monitoring and evaluating the security of such environments are limited. Specifically, there is a lack of reliable methods to automatically understand how devices behave on a network and to detect abnormal or potentially malicious activity in real time.
The aim of this project is to develop advanced artificial intelligence (AI) based methods to automatically characterise and analyse the network behaviour of IoT/OT devices. By improving the robustness and scalability of machine learning models for traffic analysis, this project seeks to strengthen cybersecurity management in large scale cyber physical environments.
You will join a dynamic research team within the School of Electrical Engineering and Telecommunications (EE&T), working closely with a senior academic, experienced researchers, and honours students. The project also involves collaboration with industry partners who are interested in deploying and testing the developed solutions in real operational networks. This provides a unique opportunity to contribute to cutting edge research while gaining hands on experience with real commercial IoT devices and real world network traffic data.
This project sits at the intersection of computer networking and applied machine learning. It aims to develop low cost, scalable, and accurate analytical tools that can be deployed in live environments to enhance cybersecurity resilience.
Eligibility: Due to the nature of the project and potential access to sensitive network environments, this position is open to domestic students (Australian citizens) only.
School
Electrical Engineering and Telecommunications
Research Area
Computer networking | TCP/IP protocols | Machine learning | Statistical data analysis
Suitable for recognition of Work Integrated Learning (industrial training)?
Yes
- Research environment
- Expected outcomes
- Supervisory team
- Reference material/links
This project will be conducted within a vibrant and collaborative research group comprising PhD students, honours students, postdoctoral researchers, and academic staff at UNSW. The team also works closely with industry partners interested in trialling and evaluating research outcomes in real operational networks.
You will have access to commercial IoT devices and real network traffic datasets (both live and pre recorded packet traces). The research environment emphasises theoretical development and practical validation, enabling your software tools and algorithms to be tested in realistic network conditions. Regular group meetings, research discussions, and mentorship from senior researchers will support your learning and research development.
- Development of: (a) novel algorithms for characterising the network behaviour of individual IoT device types, (b) methods for identifying relationships and dependencies between different IoT devices, (c) techniques for improving the interpretability and controllability of machine learning based traffic classifiers.
- Prototype software tools for analysing and classifying IoT network traffic.
- A written scientific technical report documenting research findings.
- High quality results may contribute to publications in international conferences or journals, and selected prototypes may be trialled in live operational networks through industry collaboration.
- Representatives from industry partners
- The UNSW research team has published many research articles in this area, which can be found at A/Prof. Gharakheili's website: https://www2.ee.unsw.edu.au/~hhabibi/publications.html
- A recommended starting point is our recent paper: A. Sivanathan et al, Real-Time and Trustworthy Classification of IoT Traffic Using Lightweight Deep Learning, IEEE Transactions on Network Science and Engineering: https://www2.ee.unsw.edu.au/~hhabibi/pubs/jrnl/25tnse.pdf