Deep Learning Intrusion Detection System for Smart Satellite Networks Based on Software Defined Networking

The project will focus on detecting anomalous traffic within SDN-based smart satellite networks using Deep Learning (DL). As IoT devices are growing at an exponential rate, so does the threats and vulnerabilities faced by IoT. As the world is migrating to the Fourth Industrial Revolution (4IR), there will be a heightened growth of IoT systems that will be required to maintain uninterrupted connectivity on a large scale. The need for early detection of cyber-attacks that target smart satellite networks based IoT systems is evident. Additionally, DL can analyse complex and large traffic data patterns in the least amount of time and is therefore being adopted in areas of health, mining, and agriculture amongst others. The project will contribute to SmartSat’s focus areas of Advanced Satellite Systems, Sensors and Intelligence specifically the data security of satellite systems, and the Security of Satellite-based IoT services with Deep Reinforcement Learning and is in close association with SmartSat CRC’s Project P2-36, Cyber Jeopardy and Response (CY-JAR).

P2.44s

Project Leader:
Professor Jill Slay, The University of South Australia

PhD Student:
Uakomba Uhongora, The University of South Australia

Participants: