Contactless human activity recognition and vitals sensing for next generation smart homes and healthcare centers

Farooq, Muhammad (2025) Contactless human activity recognition and vitals sensing for next generation smart homes and healthcare centers. PhD thesis, University of Glasgow.

Abstract

According to the House of Commons Library United Kingdom Parliament, approximately 7.9 million people live alone in the UK. Out of 7.9 million,over 3.1 million adults aged 65 and above live alone in the UK. The older population in the UK is projected to grow, with people aged 65 and over making up 24% of the population by 2043 (17.4 million people). Given this background, the development of a monitoring system that can recognize an emergency or health condition is desired by healthcare providers and families of individuals living alone. Unusual changes in a lonely living person’s regular daily mobility routine at home can indicate early symptoms of developing health problems.

This thesis paves the way to develop a novel system that exploit Energy, LoRa, WiFi, RF and radar based technologies to monitor human activity, including presence detection, postural transitions such as walking, sitting, standing, lying, and fall detection. Additionally, it introduces a robust framework for contactless vital signs monitoring, enabling accurate measurement of breath rate, pulse, heart rate, and heart sounds. The integration of AI-driven anomaly detection enhances the system’s ability to identify potential health risks in real-time. The research further explores the fusion of human activity recognition with vital signs monitoring to develop a complete, scalable, and privacy preserving solution for both general well-being and clinical healthcare applications. By developing advanced signal processing techniques and machine learning models, the proposed system aims to provide an efficient, non-invasive alternative to conventional health monitoring methods.

Further contributions include a contactless framework for sleep pattern recognition, utilizing micro doppler radar signals to classify sleep postures and detect abnormalities associated with autism spectrum disorder. The study also advances non-invasive health monitoring through radar systems for vital signs detection, achieving high accuracy in respiration and heart rate variability assessment. Moreover, heart sound detection and analysis enhance cardiac monitoring, improving pulse detection, heart rate estimation, and overall reliability of vital signs monitoring. This work contributes to the future of smart living by ensuring continuous, real time health monitoring without compromising user’s comfort and privacy. Future research will focus on improving system adaptability, enhancing multimodal sensing capabilities, and addressing data security challenges to facilitate widespread deployment in next generation smart homes and medical facilities.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: T Technology > T Technology (General)
• Colleges/Schools: College of Science and Engineering > School of Engineering > Autonomous Systems and Connectivity
• Supervisor's Name: Abbasi, Professor Qammer, Hasan Abbas, Dr. Hasan Abbas and Taha, Dr. Ahmed
• Date of Award: 2025
• Depositing User: Theses Team
• Unique ID: glathesis:2025-85253
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 26 Jun 2025 13:50
• Last Modified: 26 Jun 2025 14:09
• Thesis DOI: 10.5525/gla.thesis.85253
• URI: https://theses.gla.ac.uk/id/eprint/85253

Related URLs

• Article DOI
• Article DOI
• Article DOI
• Book section
• Conference item
• Conference item
• Conference item
• Conference item
• Conference item
• Conference item
• Conference item
• Conference item
• Conference item
• Conference item
• Conference item