Trustworthy, Software-Defined Cyberattack Detection and Mitigation at the Network Edge (TruSDEd)

The TruSDEd project addresses how to enable in-situ and trustworthy cybersecurity provisioning for resource-constrained  Information and Communication Technology (ICT) deployments against local and Internet-originated cyberattacks. 

Securing the IoT and, more generally, constrained ICT deployments (outside of the datacentre) that connect to the periphery of the Internet is far from a solved challenge. The ever-increasing penetration and diversity of such deployments (from smart cities to offshore and from autonomous vehicles to satellite missions) also comes with an ever-increasing list of targeted cyber-incidents that compromise them and can go undetected for prolonged periods of time.  

It is therefore clear that there is an urgent need to improve the cybersecurity characteristics of such deployments before they can have a higher uptake that will improve automation, process efficiency, and hence the digital economy.  

The team argue that improving cybersecurity should satisfy the following requirements: a) use advanced monitoring and ML/AI-based inference to profile the normal behaviour of a particular deployment so that it can detect newly-evolving cyberattacks; b) support strongly authenticated components and software modules that can be trusted to operate alongside the deployed infrastructure; and c) be provided in a cost-effective manner to be extensible, not require every potentially miniature device to be individually patched, and not depending on expensive, static, or solely Cloud-based solutions that each have their own weaknesses in protecting non-standard ICT environments.   

The research will provide the necessary algorithms, software, and hardware architecture to meet the above requirements and will be demonstrated through the development of a prototype system and hosting platform.