antifragile programmable networks over multiple defence domains and bearers
Key Points
- DAIRNet Theme: Distributed Multi-Domain Networks
- University of Adelaide
- Point of contact: A/Prof Hung Nguyen
bottom line up front
In kind of the way Starlink was able to upgrade when a threat showed up, we need to be able to have that agility.
David Tremper, Director of Electronic Warfare, the Office of the US Secretary of Defense, April 2022
Project Aim: Military communication networks that rapidly and securely adapt to changes in the environment
Outcome: Antifragile programmable tactical networks that are robust to changes and improve with external and internal stressors with software-defined networking and Machine Learning.
So what? Unprecedented capability to automatically adapt tactical comms at the speed of relevance
Next Steps: further prototype development and field testing – especially Anti-fragility capability
problem addressed
Programmable multi-domain, multi-bearer military communication systems with cutting-edge AI/ML and networking technologies – legacy and current waveforms (SIlVUS, WiFi, 5G, BATMAN): Data-driven, Fast adaptation (speed of relevance), and Improves with stressors.
Key Technologies: Software Defined Networking (SDN) and P4 , Graph Neural Network (GNN), Reinforcement Learning.
outcomes
- Novel (Sovereign) RASPSWITCH solution to allow programmable control of various bearers WiFi, 4G, BATMAN and SILVUS.
- International award-winning machine learning solutions to make rapid traffic decisions in seconds
- Deployable Prototype: We have tested the solutions in our testbed with 4 real SilVUS and 6 BATMAN radios
- Scalable – the solution was shown to scale to networks of more than 50 nodes in emulations
the big picture for Defence
here are two sides to the coin of complex communication systems. On the one hand, there are untold challenges: hidden cyber vulnerabilities, long development delays, interoperability of legacy systems, and unexpected and undesirable bugs. On the other hand, there is vast potential: seamless collaboration between military systems across services and domains → intelligent decision superiority.
We could maximise the benefits and minimise the liabilities of military communication systems by harnessing AI/ML and software-defined communication technologies.
publications
- Phu, AT, Li, B, Ullah, F, Huque, TU, Naha, R, Babar, A & Nguyen, H 2023, ‘Defending SDN against packet injection attacks using deep learning’, arXiv preprint arXiv:2301.08428. DOI: 10.48550/ARXIV.2301.08428
- Nguyen, N, Nguyen, DD, Kim, J, Rizzo, G & Nguyen, H 2022, ‘Multi-Agent Data Collection in Non-Stationary Environments’, 2022 IEEE 23rd International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM), IEEE, pp. 120-129. DOI: 10.1109/WoWMoM54355.2022.00023
- Dinani, MA, Holzer, A, Nguyen, H, Marsan, MA & Rizzo, G 2022, ‘Vehicle Position Nowcasting with Gossip Learning’, IEEE Wireless Communications and Networking Conference (WCNC), pp. 728-733. DOI: 10.1109/WCNC51071.2022.9771706
