![](https://australiancybersecuritymagazine.com.au/wp-content/uploads/2022/07/img_Digital-automation.jpg)
Digital automation promises to accelerate productivity at a massive scale, but with networked digital infrastructure comes the risk of unexpected faults or malign activities due to undetected network vulnerabilities.
To solve this issue, such systems will need to detect and respond to unusual patterns of activity.
University of Southern Queensland PhD researcher Chris Davey and project leader Professor Ravinesh Deo have been hard at work on this task as part of the Australian Department of Defence’s Artificial Intelligence for Decision Making (AI4DM) initiative.
The challenge, posed by Defence cybersecurity researchers, supports work developing machine learning tools to detect patterns of events in information sent over networks for the purpose of defence cyber applications. And in particular, system monitoring.
The AI4DM is a highly competitive initiative where successful applicants are awarded $30,000 to identify and propose novel solutions to the challenges of significant importance to the Australian defence and national security community.
“Non-standard network protocols often require reverse engineering based on only a few example network communications, which requires a great deal of skill, is time-consuming and expensive,” Mr Davey said.
“Our AI4DM response is focused on determining the feasibility of training a machine learning model which can extract protocol structures and to provide a mechanism for the model to be updated for new protocols as part of its overall operational lifecycle.”
Mr Davey said the research challenge project was a small part of a broader project which aims at identifying potentially malign or disruptive network events.
“It is grounded within the cybersecurity domain and aims to extract useful information about the structure and type of data from often proprietary network protocols,” he said.
“Such information is used as part of a larger data processing pipeline that characterises activities and events occurring within the networked control system.
“The project seeks to achieve this by developing a data driven methodology for the training of a byte-level sequence-to-sequence neural network model.
“If successful, the developed methodology, documentation and demonstration capability will contribute to the larger project of our collaborators within Defence whose aim is to automatically monitor and identify potentially adversarial events both historically and online.”
Mr Davey said he was honoured to participate in the AI4DM initiative and to spearhead the exciting yet challenging work.
“This is a fantastic opportunity to collaborate with the Defence team and contribute towards their larger research objectives,” he said.
“The ability to automatically characterise the overall state of a networked command and control system will be highly beneficial for pro-active operational monitoring as well as post-event fault diagnosis.”
Professor Deo said the cybersecurity sector was one of the most rapidly growing industries globally.
“Protecting personal and corporate data through cyber security is paramount to some of the world’s biggest industries,” he said.
“Developing novel artificial intelligence methods and exploring more intelligent ways to redefine defence cybersecurity through this AI4DM national challenge demonstrates the University’s growing research footprint in this field.”
This research is funded by the Australian Government under the Artificial Intelligence for Decision Making (AI4DM) Initiative.
Mr Davey is currently pursuing a three-year PhD project in AI-enabled communications under supervision of Professor Ravinesh Deo, Adjunct Professor Ismail Shakeel, Professor Jeffrey Soar and Adjunct Professor Sancho Salcedo-Sanz. This PhD project is supported by a postgraduate scholarship funded by the Department of Defence.