Intrusion detection

Download @github


A self-configurable intrusion detection system for detecting cyber-physical attacks in smart home and IoT environments. MAGPIE takes in consideration both cyber and physical sources of data, as well as presence inference to dynamically optimise attack detection at runtime.

Computational offloading


Automated coordination of computational offloading in the Internet of Things. MEDICI employs mathematical modelling and machine learning for accurately estimating the task completion time and energy at different locations. It is currently being validated in a Smart Factory.



Dashboard helping a security operations centre to prioritise the reports of suspected social engineering security incidents in an organisation. It achieves this based on machine learning that predicts the reliability of each report on a given platform based on the platform usage patterns and other user characteristics. 

ISEC Research Group


University of Greenwich,

Old Royal Naval College,

London, UK


Tel: +44 (0)20 8331 9612

email: g.loukas at