Detection of heatbeats in wireless signal Detection of heartbeats in wireless signal

Date
2006
Authors
Zhou, Qin, 1980
Contributor
Advisor
Department
Instructor
Depositor
Speaker
Researcher
Consultant
Interviewer
Annotator
Journal Title
Journal ISSN
Volume Title
Publisher
Volume
Number/Issue
Starting Page
Ending Page
Alternative Title
Abstract
Doppler radar remote detection of heart activity is a promising technique for unobtrusive health monitoring and life sensing. However, so far this approach has been limited to practical use. The difficulty lies in distinguishing targeted heartbeat signals from artifacts produced by the presence of other people or environmental motion. We are exploring separation of signals from multiple subjects in single and multiple antenna systems. To achieve this objective, the first thing is to detect how many subjects exist in a given area. This thesis mainly focus on design of optimal detectors for this problem. Currently, we are only aim to distinguish among the presence of 2, 1, or 0 subjects with one or two antennas. For this particular application, the primary concern of a good detector is the selection of an optimality criterion and a good model of the data to statistically describe the principle features. The wireless heartbeat signal due to Doppler shift is a very weak nonstationary signal with time-variant periods. Within each detection window, it is modeled as a sinusoid with constant frequency, but staircase jumpily variant phase and magnitude from subwindow to subwindow. Detection turns into determining how many such sinusoids are present. For detection problem with so many unknown parameters, General Likelihood Ratio Test (GLRT) is presented as solution. Although there is no optimality associated with the GLRT, in practice, it appears to work quite well. We have shown that it is possible to distinguish among the presence of 2, 1, or 0 subjects, even with single antenna. The key of success is to find good window structure and length. The detection probability can be improved by using more antennas. As is well known, GLRT need Maximum Likelihood Estimation (MLE) of unknown parameters, especially MLE of the heartbeat frequency. For single person or nobody present cases, MLE of the frequency is equivalent to find the peak of the averaged FFT over subwindows. Based on FFT and GLRT, fast novel heartbeat detection algorithms are presented and evaluated for both real and complex data model. Although real data model can use SVD combination of multiple antennas to improve performance, complex data model still outperforms the former scheme. Some good window structures are recommended, and their simulation results show that the presented complex detector has a very excellent performance.
Description
Thesis (M.S.)--University of Hawaii at Manoa, 2006.
Includes bibliographical references (leaves 78-82).
xiii, 82 leaves, bound ill. 29 cm
Keywords
Heart rate monitoring -- Equipment and supplies
Citation
Extent
Format
Geographic Location
Time Period
Related To
Theses for the degree of Master of Science (University of Hawaii at Manoa). Electrical Engineering; no. 4152
Table of Contents
Rights
All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner.
Rights Holder
Local Contexts
Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.