Date of Award

Fall 1996

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering - (Ph.D.)

Department

Electrical and Computer Engineering

First Advisor

Stanley S. Reisman

Second Advisor

Peter Engler

Third Advisor

Joseph Frank

Fourth Advisor

Bonnie K. Ray

Fifth Advisor

Leon Cohen

Abstract

Recently, joint time-frequency signal representation has received considerable attention as a powerful tool for analyzing a variety of signals and systems. In particular, if the frequency content is time varying as in signals of biological origin which often do not comply with the stationarity assumptions, then this approach is quite attractive. In this dissertation, we explore the possibility of better representation of two particular biological signals, namely heart rate variability (HRV) and respiration. We propose the use of time-frequency analysis as a new and innovative approach to examine the physical and mental exertion attributed to exercise. Two studies are used for the main investigation, the "preliminary" and "anticipation" protocols.

In the first phase of this work, the application of five different bilinear representations on modeled HRV test signals and experimental HRV and respiration signals of the preliminary protocol is evaluated. Each distribution: the short time Fourier transform (STFT), the pseudo Wigner-Ville (WVD), the smoothed pseudo Wigner-Ville (SPWVD), The Choi-Williams (CWD), and the Born-Jordan-Cohen (RID) has unique characteristics which is shown to affect the amount of smoothing and the generation of cross-terms differently . The CWD and the SPWVD are chosen for further application because of overcoming the drawbacks of the other distributions by providing higher resolution in time arid frequency while suppressing interferences between the signal components.

In the second phase of this research, the SPWVD and CWD are used to investigate the presence of an anticipatory component due to the stressful exercise condition as reflected in the HRV signal from a change in behavior in the autonomic nervous system. By expanding the concept of spectral analysis of heart rate variability (HRV) into time-frequency analysis, we are able to quantitatively assess the parasympathetic (HF) and sympatho-vagal balance (LF:HF) changes as a function of time. As a result, the assessment of the autonomic nervous system during rapid changes is made.

A new methodology is also proposed that adaptively uncovers the region of parasympathetic activity. It is well known that parasympathetic activity is highly correlated with the respiration frequency. This technique traces the respiration frequency and extracts the corresponding parasympathetic activity from the heart rate variability signal by adaptive filtering.

Share

COinS