EPSRC logo

Details of Grant 

EPSRC Reference: EP/Z00084X/1
Title: IntentMAPS: Anticipating Intent facilitated by Multi-scale Adaptive Passive RF Sensing
Principal Investigator: McCann, Professor J
Other Investigators:
Yap, Professor M Crisp, Dr M Birch, Dr PM
Zecca, Professor M Gan, Dr L Davis, Dr GJ
Researcher Co-Investigators:
Project Partners:
University of Glasgow
Department: Computing
Organisation: Imperial College London
Scheme: Standard Research - NR1
Starts: 01 July 2024 Ends: 30 June 2026 Value (£): 1,254,299
EPSRC Research Topic Classifications:
Digital Signal Processing Networks & Distributed Systems
EPSRC Industrial Sector Classifications:
Communications Information Technologies
Related Grants:
Panel History:  
Summary on Grant Application Form
IntentMAPS arose from the Novel Sensing for UK Defence and Security sandpit to develop new sensor technology to meet the needs of defence and security. Three specific presentations informed the creation of this project, and the common requirements of these scenarios were that the technologies be non-intrusive to the general public, accurate, and importantly, cost-effective.

Taking a multi-disciplinary approach, IntentMAPS will investigate how next generation RF communications signals can be exploited to detect social primitives and interactions which might be characteristic of malicious intent in crowded settings. The use of reflected RF signals, as opposed to more common CCTV, allows the potential for new 'see-through' functionality, as well as an element of privacy preservation, as conventional, human recognisable 'images' are not formed. That is, the key advantage of RF is that it provides a 'visibility' that optical cameras cannot, i.e. it can still sense the person's activities when the individual is covered, hiding behind internal walls, and other occlusions. Since 6G is planned for wide deployment, pervasive coverage of crowded areas should be achievable, and by using reflected signals from the body, even people without wireless devices can be detected.

Combining innovations in RF sensor technologies, signal processing, social-cognition research with transformer-enhanced AI, we aim to explore the extent to which we can provide a new sensing system that achieves automated prediction of imminent terror attacks. This new sensor has the potential to provide crucial 'first look' capabilities, permitting surveillance of terrorists and other malicious actors before they adopt counter-surveillance measures, and by embedding aspects of human social cognition informed AI, earlier detection of evolving malicious acts could alert security staff in the narrow window of opportunity for intervention.

IntentMAPS aims to bring about a step change in security screening, not by a single 'moon-shot' development in any single technology, but rather by combining a series of carefully de-risked, advances across four disciplines. That is, IntentMAPS will explore the combined use of Radio Frequency (RF) sensing enhanced using Reconfigurable Intelligent Surfaces (RIS) and state of the art research on Intent Behaviours. This work has not been done before. In contrast to other comparable work, e.g. Transport for London's recent Willesden Green station Trial (using CCTV focused on tailgating, access, litter; AI could not detect aggressive behaviours due to impoverished datasets)*. IntentMAPS focuses on intent in terrorism; it builds richer, controlled data sets, avoiding broad, poorly identified categories such as 'aggression' and extends beyond vision using RIS RF sensing. To realise this adventurous and ambitious aim, we have assembled a strong multi-disciplinary team that brings together psychology, computer science/AI, signal processing and electrical engineering.

This approach, requiring a multidisciplinary team with substantial breadth of expertise, has benefitted from the unique opportunities provided by the EPSRC sandpit. New research in each of these fields is required to do this, and contributions will be made in each. Bringing all this work together presents an entirely different way to sense intent, achieving sensing beyond that of vision-based methods alone, and will contribute to a step-change understanding of RF sensing of mobile targets in crowded spaces at scale, for the first time. This is a high adventure project, where reasonably high risk is mitigated through our collective expertise and experience.

* https://foi.tfl.gov.uk/FOI-3155-2324/Smart%20Station%20End%20of%20PoC%20Presentation_Redacted.pdf

Key Findings
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Potential use in non-academic contexts
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Description This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Date Materialised
Sectors submitted by the Researcher
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
Project URL:  
Further Information:  
Organisation Website: http://www.imperial.ac.uk