DEWC Services, in collaboration with the University of Adelaide and Defence Trailblazer, is leading a research initiative into Hybrid Moving Target Defence (MTD) for container-based cloud environments, using Reinforcement Learning (RL) to create an AI-driven cyber defence system that adapts to evolving threats.
As cyber threats grow more complex, traditional static defences are no longer enough. MTD disrupts attacks by continuously altering the attack surface, making it harder for hackers to penetrate systems.
Current MTD models face challenges in scalability, cost, and real-world application. This project aims to address those issues by integrating RL, allowing for more intelligent and adaptive defence systems.
Dr Tim McKay from DEWC Services explained, “Integrating RL into Hybrid MTD improves security by enabling AI-driven optimization of MTD configurations.”
The research aims to optimize MTD deployment, ensure smooth integration with existing security systems, and develop cost-effective solutions for long-term national security. The team will test the strategies in real-world, high-risk scenarios.
Associate Professor Claudia Szabo from the University of Adelaide added, “This research will help enhance cyber defence adaptability and efficiency.”
DEWC Services will present their findings at the DTECH25 online conference on 18 March, showcasing how RL-powered Hybrid MTD can improve cloud security.
Dr Sanjay Mazumdar, Defence Trailblazer Executive Director, highlighted the project’s importance in advancing AI-driven defence solutions for national security. The team will also develop a prototype to demonstrate Hybrid MTD in real-world defence applications.
