DT Cyber will develop a novel method for analysing CPS architectures to identify which subsystems should be included in a Digital Twin platform. This will focus on the adaption of Data Flow Diagrams (DFD) to provide an effective procedure to identify cyber and physical system components, data flows, and physical interactions to identify the appropriate boundaries of DTs that can support cyber-security monitoring of CPS subsystems. We will evaluate the use of the methodology on two case studies.
DT Cyber is a joint project between Technological University of the Shannon and Queen's University Belfast.
PIs: Dr.Brian Lee (TUS)
CEASED will develop a federated anomaly detection framework to improve attack detection in smart edge hybrid information (IT) and operations technology (OT) systems such as smart manufacturing. It will use a combination of machine learning models to detect anomalous usage of attack techniques on endpoints and train the models collaboratively using federated learning.
ProvE will develop a generic trusted task execution framework to facilitate diverse Decentralised Applications (DApps) in implementing application specific Proof of X functions such that these DApps can provide attestation for their services in untrusted environments. This project will use confidential computing and Blockchain smart contracts to design the framework and protocols.
This project is to develop a solution for transmitting the video from the camera at the back of the trailer to the driver’s cab over wireless channels. The design will consider different road, environmental and weather conditions.
Malicious Domain Hunting using Graph Clustering MethodsThis project is to develop a set of malicious domain hunting algorithms by collaborating with an industrial partner. A hierarchical attributed-based graph clustering algorithm is developed to detect the domains that shared similar behaviours. It allows further assessing the risk level of unknown domains if their behaviours correlate to known malicious domains. In addition, a time-series-based detection algorithms is developed to identify outliers from domain clusters, to refine the prediction accuracy.
RobotLedger – A Decentralised Ledger System for Multiple Robot CollaborationsMulti-robot systems have wide application scenarios, especially in dangerous, unknown, or hazardous environments, e.g. humanitarian demining, search and rescue, underwater exploration, and surveillance. The consensus problem is fundamental in multi-robot coordination to achieve a common view on the world or to converge to a single decision. This project is to develop a blockchain based platform for multi-robot to complete a task collaboratively in untrusted environments, e.g. the robot may be hacked or the sensors may be blurred.
Distributed Deep Learning System for Privacy Preserving Video Retrieval in Large Scale Visual SensingA large number of surveillance cameras have been deployed by local business or individuals in cities. The generated footages can facilitates abundant applications, e.g. for self-driving cars to query traffic status around the corner, for police to investigate crime scenes, and for local authorities to identify the problems in the city. However, privacy concerns have been the key obstacle that hinders the development of these applications. Thus, it is crucial to create a ‘Data Fence’ that can facilitate different parties to access the video with protected privacy so that only the concerned video clips, objects in the clips or extracted information are provided to the parties. The project is to develop a GDPR compliant video analysis and retravel solution for CCTV cameras and as well as a chain-of-evidence solution for trusted video footage sharing.
Distributed Analytics for a Large Urban CCTV System
This project complements to the previous project, focusing on secure storage and access control of the video footages at the cameras. It is to develop a GDPR compliant CCTV systems supporting sensitive object detection and protection, and correspondent multi-level video encryption, storage and access at the camera devices
Monetized Information-Centric Transmission Control
The project is to develop a network resource monetization scheme to incentivize the network resource sharing between edge devices. Blockchain and smart contract technologies are integrated into the Information Centric Networking (ICN) architecture to enable the accounting and payment for content delivery. It is for the edge environments contain a large number of devices from different owners, e.g. the electric autonomous driving vehicles and the smart cameras along the street, with the assistance of the 5G / 6G technologies which will enable direct device-to-device communications without involving network infrastructure.
NAPSS – National Autonomous Pod Sharing Service
NAPSS addresses the challenge of reducing CO2 emission caused by private car usages. It is to develop a mobile-on-demand service using lightweight autonomous electric pods to reduce the usage of cars. The autonomous roving ability enable the pods to pick up clients from pre-defined locations. The shared mobility function allows to maximize the pod usage by planning the routes based on the location of the clients.
VidSDN – Application-aware Video Distribution using Software Defined Networking
VidSDN is to provide a solution for the application provider and the network operator to collaborate to maximize the application performance. It develops an application aware content distribution platform using Information Centric Networking and Software Defined Networking technologies, focusing on video distribution and big data processing scenarios. The application providers can interact with network operators to to dynamically change the application topology and re-allocate network resources according to various factors, e.g. their customer locations and numbers.
Technological University of the Shannon,Athlone Campus,University Road, Athlone,Co. Westmeath.
T: 353 (090) 6468000E: email@example.com
Connect with Us