Description of field of research:

Internet of Things (IoT) devices such as connected-cameras, door-sensors, smart-lights, and smoke-alarms sourced from a diversity of vendors and deployed in large numbers create cyber-security risks at unprecedented scale for emerging smart environments (homes, buildings, enterprises, cities) and their networks. Current methods for evaluating the security posture of such environments is at best ad-hoc, and real-time monitoring and control of their behaviour on the network are lacking. Our research group has pioneered formal models empowered by artificial intelligence to profile the behaviour of IoT devices. Our aim is to develop systematic methods for both modelling and quantifying cyber risks of large-scale IoT systems, and you will be part of this exciting journey.

This project is at the cutting-edge of research into IoT Network Security, Programmable Networks, and applied Machine Learning for network traffic inference. While several tools and databases of threat-intelligence and vulnerabilities already exist in the form of open-source or vendor-specific solutions, they are often generic (dominated by IT-specific devices), unstructured, or expensive. This project is world-leading in developing solutions that couple dynamic traffic processing techniques with systematic inference models to demonstrate low-cost, scalable, and accurate analytics. Outcomes of this project are expected to lead to publications in international conferences and journals, while also developing prototypes that can be deployed in live networks.

Research Area

Cybersecurity |
Machine Learning |
Computer Networks |

This project will be carried out in a vibrant group that includes not just PhD and honours-thesis students at UNSW, but also commercial personnel from UNSW spin-out CyAmast that is building truly disruptive network cybersecurity solutions. You will get to play with live network traffic, and your solutions will be tested and deployed in real operational networks.

Expected outcomes include: (a) analysis of traffic traces from the UNSW campus network to identify certain patterns in network behaviour of IoT devices; (b) development of new databases and techniques for real-time risk assessment of network behavioural patterns; (c) prototyping and evaluation of performance and scalability trade-offs using real systems and live traffic streams.

The UNSW research team has written many research articles which can be found at Dr. Gharakheili's website: https://www2.ee.unsw.edu.au/~hhabibi/publications.html. A recommended starting point is our paper: F. Loi, A. Sivanathan, H. Habibi Gharakheili, A. Radford and V. Sivaraman, Systematically Evaluating Security and Privacy for Consumer IoT Devices, ACM IoT Security and Privacy (IoT S&P), Texas, USA, Nov 2017.