University of Hawaii

Department of Electrical Engineering

Distortion Reduction and Signal Estimation in Doppler Radar Physiological Monitoring Systems

Date: 2013-10-17
Time: 9 am
Location: Holmes 485
Speaker: Ehsan Yavari, Advisor: Dr. Olga Boric-Lubecke

Cardiopulmonary monitoring can serve as good prognostics and diagnostics tool for respiratory and cardiac ailments which are the leading cause of death in United States. Respiratory rate, heart rate, pulse pressure, and volumetric measurement are important indicators of human health status.  For measuring vital signs Doppler radar can be used to detect small movements on the chest walls induced by heart and lungs. Being a non-invasive and non-contact method, numerous researches were conducted for health monitoring such as: sleep monitoring, home healthcare monitoring, tumor tracking, baby monitoring, life detection under the rubble, and occupancy detection. Although Doppler radar is a well-developed field, physiological monitoring is a rather new application bringing its own challenges. Issues with low frequency of the physiological motion, quadrature channel imbalance, local oscillator leakage, DC offset and flicker noise limit the sensitivity and accuracy of the radar system. The goal of this work is to develop new hardware and software solutions to address challenges in Doppler radar physiological monitoring that currently limit practical applications. In particular, this work focused on improving system performance by reducing the effects of distortion. New signal processing algorithms are explored to condition the received signal for better physiological motion extraction, and estimation of life signs. Numerous experiments with mechanical target simulating chest motions, and human subjects have been done to explore the effectiveness of the proposed methods.