AARNet (APR – 3005)

Using AI to Predict Network Faults and Cyber Security Events

Engineering, IT, Mathematics and Statistics

ABOUT THE INDUSTRY PARTNER

The Australian Academic and Research Network (AARNet) is Australia’s national research and education network, connecting universities, research institutions, schools, hospitals and cultural organisations through high-performance digital infrastructure.

AARNet delivers advanced networking, cyber security and digital services that enable world-leading research and collaboration across Australia. Working alongside experienced engineers and technology specialists, you’ll contribute to innovative projects that improve the reliability, resilience and performance of one of Australia’s most important research networks.

WHAT’S IN IT FOR YOU?

This internship offers an exciting opportunity to apply machine learning, artificial intelligence and data science techniques to solve complex challenges in network engineering and cyber security.

During the internship, you’ll have the opportunity to: 

• Apply your research to large-scale operational network datasets. 
• Explore innovative AI techniques to predict network faults and security events. 
• Work alongside experienced network engineers and technology specialists. 
• Develop practical experience translating research into operational solutions. 
• Contribute to improving the resilience and reliability of Australia’s national research network. 

RESEARCH TO BE CONDUCTED

AARNet has a need to merge many different disparate sources of information about the state of its network in a way that makes sense to engineers and support staff, enabling proactive engineering and security responses. 
 
The primary goal of this project is to raise early awareness of possible upcoming network events by identifying abnormalities through heuristic analysis of historical network state. Working with AARNet, the successful applicant will investigate the relationships between network equipment telemetry, sampled network traffic flows, historical records and other forewarning indicators to develop a data model that can be applied using data science and artificial intelligence (AI) approaches to predictively raise awareness of an impending network fault or security event. 
 
Potential applications of the research include predictive capacity expansion alerting, identifying failing fibre optic and other network connections, detecting capacity exhaustion on routers and firewalls during burst events, and improving visibility of the true costs associated with particular network behaviours. 
 
The project may involve analysing data from network monitoring systems, equipment telemetry, configuration management database (CMDB) sources, NetFlow and other traffic flow sampling data, as well as external data sources such as weather and emergency services reports. Security events may also be investigated by detecting early exploratory probes of the network that could indicate a probable denial-of-service attack. 
 
The successful applicant will work with AARNet to determine the most appropriate analytical approach for the project. Depending on the candidate’s interests and expertise, this may involve machine learning (ML), neural networks, graph models, digital twins, large language models (LLMs), or other AI techniques to explore historical records and support network operators through predictive decision-making. 

SKILLS WISH LIST

If you’re a postgraduate research student and meet some or all the below we want to hear from you. We strongly encourage women, indigenous and disadvantaged candidates to apply:

Essential 

• Strong background in mathematics, statistics, computer science or a related discipline. 
• Experience analysing complex datasets. 
• Knowledge of machine learning or artificial intelligence. 
• Strong analytical and problem-solving skills. 
• Excellent written and verbal communication skills. 

Desirable 

• Graph analytics. 
• Network engineering or telecommunications. 
• Cyber security. 
• Python or similar data science programming languages. 
• Neural networks or deep learning. 
• Large language models (LLMs). 

RESEARCH OUTCOMES

By the end of the internship, the successful applicant will have contributed towards: 

• A data model that supports early awareness of potential network faults and security events. 
• Improved understanding of relationships between network telemetry, traffic flows, historical records and other operational data sources. 
• Recommendations for applying AI and data science techniques to support predictive network operations. 
• Prototype models or proof-of-concept tools that assist engineers and support staff in identifying emerging operational issues.

ADDITIONAL DETAILS

The intern will receive $3,300 per month of the internship, usually in the form of scholarship payments.

It is expected that the intern will primarily undertake this research project during regular business hours and maintain contact with their academic mentor throughout the internship either through face-to-face or phone meetings as appropriate.

The intern and their academic mentor will have the opportunity to negotiate the project’s scope, milestones and timeline during the project planning stage.

Please note, applications are reviewed regularly and this internship may be filled prior to the advertised closing date if a suitable applicant is identified. Early submissions are encouraged.

INTERNSHIP CONTACT

CONNECT WITH APR.INTERN