Thesis

Enhancing livestock and human health monitoring via analysis of electronic sensor data

Downloadable Content

Download PDF
Creator
Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2018
Thesis identifier
  • T14882
Person Identifier (Local)
  • 201091315
Qualification Level
Qualification Name
Department, School or Faculty
Abstract
  • This thesis presents a body of work involving novel algorithms for enhancing the effectiveness of low-cost sensors in monitoring applications. A holistic approach has been taken in this work in that modelling, simulation, and monitoring tools have been developed from scratch with a number of novel ideas. As its first contribution, the thesis presents a new simulation tool, WSNSIM - a tool for performance analysis of wireless sensor networks (WSN) formed by sensor nodes deployed on farm animals for monitoring of health and oestrus. In this application of wireless sensor networks, the mobility and herding patterns are modelled using statistical tools such as the Gamma density function, mean index of adequacy (MIA), exponential distribution, K-means clustering etc. to give rise to network simulation that is based on accurate herd behaviour. The simulation results are used in evaluation of novel protocol ideas customized to the needs of farm monitoring. The paper [153] presents a new simulation tool for performance analysis of wireless sensor networks (WSN) deployed on farm animals. The second and third key contributions of the thesis investigate monitoring of human body joints for the purpose of gait and upper limb motion assessment. Unlike the standard approach of marker-based joint monitoring for motion assessment, this work investigates the viability of the Kinect sensor for joint motion monitoring. To this end, two novel tools are developed that incorporate statistical, image processing and computer vision algorithms. The first tool GLSKEL is an intuitive 3D interface and kinematics model for continuous motion capture and analysis of human gait that can be useful for clinical practitioners. The second tool, JAFAKEC helps with tracking and calculation of joint angles based on point cloud data. This functionality can be very helpful for monitoring of the gait and arm motions of mobility-impaired patients using the Kinect sensor. This thesis also details the mathematical methods and algorithms applied on the point cloud to improve accuracy of the joint angle calculation.
Advisor / supervisor
  • Stankovic, Lina
  • Andonovic, Ivan
Resource Type
DOI
Date Created
  • 2018
Former identifier
  • 9912600388802996

Relations

Items

Thumbnail Title Date Uploaded Visibility Actions
file details PDF of thesis T14882 2021-07-02 Public Download