Design of software defined radio based testbed for smart healthcare

Ashleibta, Aboajeila Milad Abdulhadi (2022) Design of software defined radio based testbed for smart healthcare. PhD thesis, University of Glasgow.

Full text available as:
[img] PDF
Download (17MB)


Human Activity Recognition (HAR) help to sense the environment of a human being with an objective to serve a diverse range of human-centric applications in health care, smart-homes and the military. The prevailing detection techniques use ambient sensors, cameras and wearable devices that primarily require strenuous deployment overheads and raise privacy concern as well. Monitoring human activities of daily living is a possible way of describing the functional and health status of a human. Therefore, human activity recognition (HAR) is one of genuine components in personalized life-care and healthcare systems, especially for the elderly and disabled. Recent advances in wireless technologies have demonstrated that a person’s activity can modulate the wireless signal, and enable the transfer of information from a human to an RF transceiver, even when the person does not carry a transmitter. The aim of this PhD project is to design a novel, non-invasive, easily deployable, flexible and scalable test-bed for detecting human daily activities that can help to assess the general physical health of a person based on Software Defined Radios (SDRs). The proposed system also allows us to modify the power level of transceiver model, change the operating frequency, use self-design antennas and change the number of subcarriers in real-time. The results obtained using USRP based wireless sensing for activities of daily living are highly accurate as compared to off-the-shelf wireless devices each time when activities and experiments are performed. This system leverage on the channel state information (CSI) to record the minute movement caused by breathing over orthogonal frequency division multiplexing (OFDM) in multiple sub-carriers. The proposed system combines subject count and activities performed in different classes together, resulting in simultaneous identification of occupancy count and activities performed. Different machine learning algorithms namely K-Nearest Neighbour, Decision Tree, Discriminant Analysis, and Naıve Bayes are used to evaluate the overall performance of the test-bed and achieved a high accuracy. The K-nearest neighbour outperformed all classifiers, providing an accuracy of 89.73% for activity detection and 91.01% for breathing monitoring. A deep learning convolutional neural network is engineered and trained on the CSI data to differentiate multi-subject activities. The proposed system can potentially fulfill the needs of future in-home health activity monitoring and is a viable alternative for monitoring public health and well being.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: T Technology > TA Engineering (General). Civil engineering (General)
Colleges/Schools: College of Science and Engineering > School of Engineering
Supervisor's Name: Imran, Professor Muhammad and Abbasi, Dr. Qammer
Date of Award: 2022
Depositing User: Theses Team
Unique ID: glathesis:2022-83206
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 20 Oct 2022 13:51
Last Modified: 20 Oct 2022 13:54
Thesis DOI: 10.5525/gla.thesis.83206
Related URLs:

Actions (login required)

View Item View Item


Downloads per month over past year