Research units A-Z

Read the white paper: https://eprints.gla.ac.uk/395612/

Researchers from the Centre for Integrated Sensing, Communication and Computing for Cognitive Cities (ISAC³) have published two studies addressing the reliability and security of future intelligent wireless environments. These contributions support the Centre’s vision of cities that can sense, understand, compute and respond intelligently to the needs of people, infrastructure and the environment.

Improving communication and localisation under electromagnetic interference

The paper “Integrated EMI-Aware RIS Framework for Robust Communication and Localization: Simulation and Experimental Evaluation” was published on 1 July 2026 in the IEEE Journal of Selected Topics in Electromagnetics, Antennas and Propagation.

The study develops an electromagnetic interference-aware framework for reconfigurable intelligent surfaces. Through simulation and experimental evaluation, the proposed framework demonstrates how an RIS can be intelligently configured to mitigate unwanted interference and improve the reliability of wireless communication and localisation.

This work contributes to the ISAC³ vision by supporting adaptive wireless environments that can respond to changing electromagnetic conditions. Such resilience will be essential for dependable communication and localisation services within future Cognitive Cities.

Reference: S. Hassouna, A. A. Nasir, J. U. R. Kazim, M. Zubair, M. A. Imran and Q. H. Abbasi, “Integrated EMI-Aware RIS Framework for Robust Communication and Localization: Simulation and Experimental Evaluation,” IEEE Journal of Selected Topics in Electromagnetics, Antennas and Propagation, vol. 2, pp. 213–233, 2026.

Strengthening cyber-physical security in RIS-enabled ISAC systems

The paper “RISCS: RIS-Aided Integrated Sensing and Communications – A Cyber-Physical Security Perspective” was published online on 13 March 2026 in IEEE Transactions on Consumer Electronics.

The study examines cyber-physical threats affecting RIS-assisted communication and RIS-enabled integrated sensing and communication systems, including jamming, spoofing, eavesdropping and data-injection attacks. It introduces a secure RIS-enabled ISAC framework and experimentally validates an anomaly-detection approach using a pilot-aided OFDM testbed and 4,096 RIS configurations.

The proposed method jointly analyses received signal quality and channel-estimation behaviour to identify potential eavesdroppers without prior knowledge of their locations. This research contributes to the ISAC³ vision by addressing the security and trustworthiness of the shared sensing, communication and computing infrastructure required by future Cognitive Cities.

Reference: S. Hassouna, M. Zubair, A. Raza, M. Ahmed, J. Kazim, M. A. Imran and Q. H. Abbasi, “RISCS: RIS-Aided Integrated Sensing and Communications – A Cyber-Physical Security Perspective,” IEEE Transactions on Consumer Electronics, early online publication, 13 March 2026.

Together, these publications address two important requirements for future Cognitive Cities: resilience against electromagnetic interference and protection from cyber-physical threats. They demonstrate how intelligent and programmable wireless environments can provide more reliable, secure and adaptive sensing, communication and localisation services.

Researchers from the Centre for Integrated Sensing, Communication and Computing for Cognitive Cities have published three studies advancing intelligent healthcare through contactless radio-frequency sensing, edge artificial intelligence and medical-image analysis.

Together, these contributions demonstrate how sensing and computing technologies can support earlier disease detection, continuous monitoring and more accessible healthcare. This closely supports the Centre’s vision of creating intelligent environments that can sense, understand and respond to the health and wellbeing needs of individuals and communities.

Real-time RF vital-sign monitoring at the edge

The paper “EdgeCaps: Pruned Capsule Networks for RF Vital Sign Measurement in Real-time” was published online on 26 May 2026 in IEEE Transactions on Artificial Intelligence.

RF sensing provides a contactless way of monitoring vital signs such as respiration. However, many existing deep-learning models require substantial computing resources, limiting their use on low-cost and portable devices.

The researchers developed EdgeCaps, a compact artificial-intelligence framework that combines structured model pruning and knowledge distillation to enable real-time RF vital-sign monitoring on resource-constrained edge devices. The proposed model reduces computational complexity while maintaining high measurement accuracy and was experimentally demonstrated through continuous respiration monitoring on a Raspberry Pi.

The study advances the Centre’s vision by bringing sensing and intelligence closer to the user. Processing health information locally can reduce communication overhead and support scalable, energy-efficient and privacy-aware monitoring in homes, hospitals and assisted-living environments.

AI-assisted early detection of glaucoma

The paper “Hybrid ResNet–Vision Transformer Ensemble for Early Glaucoma Detection from Fundus Images” was published online on 12 January 2026 in IEEE Sensors Journal.

Glaucoma is a progressive eye disease that can cause irreversible vision loss when it is not identified at an early stage. The study introduces a hybrid deep-learning framework combining a ResNet50 model with a Vision Transformer to analyse retinal fundus images.

The two architectures provide complementary capabilities: ResNet50 identifies detailed local features within retinal images, while the Vision Transformer captures broader contextual relationships. The combined model demonstrated strong glaucoma-classification performance across a large and diverse fundus-image dataset.

This work supports the Centre’s wider health-monitoring objectives by demonstrating how intelligent sensing and advanced computing can assist early clinical screening. Such technologies could improve access to diagnostic support, particularly in communities where specialist ophthalmology services are limited.

Defining the future of contactless RF health sensing

The review paper “Non-Invasive Vital Signs Sensing: Advances, Challenges, and Future Directions in Radio Frequency Based Techniques” was published online on 12 November 2025 in the IEEE Open Journal of Antennas and Propagation.

The paper provides a comprehensive review of radar and Wi-Fi technologies for contactless monitoring of respiration rate, heart rate and other physiological activity. It examines the underlying sensing principles, signal-processing methods, machine-learning techniques and performance of current RF-based monitoring systems.

The study also identifies important challenges that must be addressed before widespread deployment, including body movement, multiperson monitoring, electromagnetic interference, data privacy, security and the availability of representative datasets. It outlines future research directions involving edge computing, intelligent signal processing and lightweight machine-learning models.

By establishing the current state of RF healthcare sensing and identifying the remaining technological challenges, the paper provides an important foundation for developing dependable contactless monitoring systems for future Cognitive Cities.

Supporting healthier and more responsive Cognitive Cities

Collectively, these publications connect three important areas of future healthcare: contactless physiological sensing, efficient intelligence at the network edge and AI-assisted medical diagnosis.

The research demonstrates how integrated sensing and computing can enable environments that continuously and unobtrusively support human health. These capabilities could contribute to remote patient monitoring, elderly care, early disease screening and personalised healthcare services within future connected communities.

References

1. P. Ishabakaki, M. Mollel, H. Hameed, H. Abbas, M. Imran and Q. H. Abbasi, “EdgeCaps: Pruned Capsule Networks for RF Vital Sign Measurement in Real-time”, IEEE Transactions on Artificial Intelligence, published online 26 May 2026.
2. F. Tariq, H. Hameed, T. Malik, M. Mollel, M. A. Imran and Q. H. Abbasi, “Hybrid ResNet–Vision Transformer Ensemble for Early Glaucoma Detection from Fundus Images”, IEEE Sensors Journal, published online 12 January 2026.
3. P. A. Ishabakaki, H. Hameed, M. Farooq, M. Mollel, H. Abbas, M. A. Imran and Q. H. Abbasi, “Non-Invasive Vital Signs Sensing: Advances, Challenges, and Future Directions in Radio Frequency Based Techniques”, IEEE Open Journal of Antennas and Propagation, published online 12 November 2025.