�

� �

�

DOPPLER RADARPHYSIOLOGICALSENSING

�

� �

�

WILEY SERIES IN BIOMEDICAL ENGINEERING ANDMULTI-DISCIPLINARY INTEGRATED SYSTEMS

KAI CHANG, SERIES EDITOR

Advances in Optical Imaging for Clinical MedicineNicusor Iftimia, William R. Brugge, and Daniel X. Hammer (Editors)

Antigen Retrieval Immunohistochemistry Based Research and DiagnosticsShan-Rong Shi and Clive R. Taylor

Introduction to Nanomedicine and NanobioengineeringParas N. Prasad

Biomedical Image UnderstandingJoo-Hwee Lim, Sim-Heng Ong, and Wei Xiong (Editors)

Doppler Radar Physiological SensingOlga Boric-Lubecke, Victor M. Lubecke, Amy D. Droitcour, Byung-Kwon Park,and Aditya Singh (Editors)

�

� �

�

DOPPLER RADARPHYSIOLOGICALSENSING

Edited by

OLGA BORIC-LUBECKE

VICTOR M. LUBECKE

AMY D. DROITCOUR

BYUNG-KWON PARK

ADITYA SINGH

�

� �

�

Copyright © 2016 by John Wiley & Sons, Inc. All rights reserved

Published by John Wiley & Sons, Inc., Hoboken, New Jersey

Published simultaneously in Canada

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form orby any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except aspermitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the priorwritten permission of the Publisher, or authorization through payment of the appropriate per-copy fee tothe Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax(978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission shouldbe addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission.

Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts inpreparing this book, they make no representations or warranties with respect to the accuracy orcompleteness of the contents of this book and specifically disclaim any implied warranties ofmerchantability or fitness for a particular purpose. No warranty may be created or extended by salesrepresentatives or written sales materials. The advice and strategies contained herein may not be suitablefor your situation. You should consult with a professional where appropriate. Neither the publisher norauthor shall be liable for any loss of profit or any other commercial damages, including but not limited tospecial, incidental, consequential, or other damages.

For general information on our other products and services or for technical support, please contact ourCustomer Care Department within the United States at (800) 762-2974, outside the United States at(317) 572-3993 or fax (317) 572-4002.

Wiley also publishes its books in a variety of electronic formats. Some content that appears in print maynot be available in electronic formats. For more information about Wiley products, visit our web site atwww.wiley.com.

Library of Congress Cataloging-in-Publication Data:

Doppler radar physiological sensing / edited by Olga Boric-Lubecke, Victor M. Lubecke,Amy D. Droitcour, Byung-Kwon Park, Aditya Singh.

p. ; cm.Includes bibliographical references and index.ISBN 978-1-118-02402-7 (cloth)I. Boric-Lubecke, Olga, 1966-, editor. II. Lubecke, Victor M., editor. III. Droitcour, Amy D., editor.

IV. Park, Byung-Kwon, editor. V. Singh, Aditya, 1984-, editor.[DNLM: 1. Heart Rate. 2. Monitoring, Physiologic–methods. 3. Respiratory Rate.

4. Signal Processing, Computer-Assisted. 5. Ultrasonography, Doppler–methods. WG 140]QP113612.1′71–dc23

2015028401

Typeset in 10/12pt TimesLTStd by SPi Global, Chennai, India

Printed in the United States of America

10 9 8 7 6 5 4 3 2 1

1 2016

�

� �

�

CONTENTS

List of Contributors xi

1 Introduction 1Amy D. Droitcour, Olga Boric-Lubecke, Shuhei Yamada,and Victor M. Lubecke

1.1 Current Methods of Physiological Monitoring, 21.2 Need for Noncontact Physiological Monitoring, 3

1.2.1 Patients with Compromised Skin, 31.2.2 Sleep Monitoring, 41.2.3 Elderly Monitoring, 5

1.3 Doppler Radar Potential for Physiological Monitoring, 51.3.1 Principle of Operation and Power Budget, 61.3.2 History of Doppler Radar in Physiological Monitoring, 8References, 16

2 Radar Principles 21Ehsan Yavari, Olga Boric-Lubecke, and Shuhei Yamada

2.1 Brief History of Radar, 212.2 Radar Principle of Operation, 22

2.2.1 Electromagnetic Wave Propagation and Reflection, 232.2.2 Radar Cross Section, 242.2.3 Radar Equation, 25

�

� �

�

vi CONTENTS

2.3 Doppler Radar, 282.3.1 Doppler Effect, 282.3.2 Doppler Radar Waveforms: CW, FMCW, Pulsed, 29

2.4 Monostatic and Bistatic Radar, 322.5 Radar Applications, 35

References, 36

3 Physiological Motion and Measurement 39Amy D. Droitcour and Olga Boric-Lubecke

3.1 Respiratory System Motion, 393.1.1 Introduction to the Respiratory System, 393.1.2 Respiratory Motion, 403.1.3 Chest Wall Motion Associated with Breathing, 433.1.4 Breathing Patterns in Disease and Disorder, 43

3.2 Heart System Motion, 443.2.1 Location and Gross Anatomy of the Heart, 453.2.2 Electrical and Mechanical Events of the Heart, 463.2.3 Chest Surface Motion Due to Heart Function, 483.2.4 Quantitative Measurement of Chest Wall Motion Due to

Heartbeat, 503.3 Circulatory System Motion, 53

3.3.1 Location and Structure of the Major Arteries and Veins, 543.3.2 Blood Flow Through Arteries and Veins, 553.3.3 Surface Motion from Blood Flow, 563.3.4 Circulatory System Motion: Variation with Age, 57

3.4 Interaction of Respiratory, Heart, and Circulatory Motion at the SkinSurface, 58

3.5 Measurement of Heart and Respiratory Surface Motion, 583.5.1 Radar Measurement of Physiological Motion, 593.5.2 Surface Motion Measurement of Respiration Rate, 593.5.3 Surface Motion Measurement of Heart/Pulse Rate, 61References, 63

4 Physiological Doppler Radar Overview 69Aditya Singh, Byung-Kwon Park, Olga Boric-Lubecke, Isar Mostafanezhad,and Victor M. Lubecke

4.1 RF Front End, 704.1.1 Quadrature Receiver, 734.1.2 Phase Coherence and Range Correlation, 774.1.3 Frequency Choice, 794.1.4 Antenna Considerations, 804.1.5 Power Budget, 80

�

� �

�

CONTENTS vii

4.2 Baseband Module, 834.2.1 Analog Signal Conditioning and Coupling Methods, 834.2.2 Data Acquisition, 85

4.3 Signal Processing, 864.3.1 Phase Demodulation, 864.3.2 Demodulated Phase Processing, 87

4.4 Noise Sources, 904.4.1 Electrical Noise, 904.4.2 Mechanical Noise, 92

4.5 Conclusions, 92References, 93

5 CW Homodyne Transceiver Challenges 95Aditya Singh, Alex Vergara, Amy D. Droitcour, Byung-Kwon Park,Olga Boric-Lubecke, Shuhei Yamada, and Victor M. Lubecke

5.1 RF Front End, 955.1.1 Single-Channel Limitations, 965.1.2 LO Leakage Cancellation, 1035.1.3 IQ Imbalance Assessment, 109

5.2 Baseband Module, 1135.2.1 AC and DC Coupling, 1135.2.2 DC Canceller, 114

5.3 Signal Demodulation, 1185.3.1 DC Offset and DC Information, 1185.3.2 Center Tracking, 1255.3.3 DC Cancellation Results, 130References, 134

6 Sources of Noise and Signal-to-Noise Ratio 137Amy D. Droitcour, Olga Boric-Lubecke, and Shuhei Yamada

6.1 Signal Power, Radar Equation, and Radar Cross Section, 1386.1.1 Radar Equation, 1386.1.2 Radar Cross Section, 1406.1.3 Reflection and Absorption, 1416.1.4 Phase-to-Amplitude Conversion, 141

6.2 Oscillator Phase Noise, Range Correlation and Residual Phase Noise, 1436.2.1 Oscillator Phase Noise, 1436.2.2 Range Correlation and Residual Phase Noise, 147

6.3 Contributions of Various Noise Sources, 1516.3.1 Phase Noise, 1516.3.2 Baseband 1/f Noise, 1546.3.3 RF Additive White Gaussian Noise, 154

�

� �

�

viii CONTENTS

6.4 Signal-to-Noise Ratio, 1556.5 Validation of Range Correlation, 1576.6 Human Testing Validation, 158

References, 168

7 Doppler Radar Physiological Assessments 171John Kiriazi, Olga Boric-Lubecke, Shuhei Yamada, Victor M. Lubecke,and Wansuree Massagram

7.1 Actigraphy, 1727.2 Respiratory Rate, 1767.3 Tidal Volume, 1797.4 Heart Rates, 1847.5 Heart Rate Variability, 1857.6 Respiratory Sinus Arrhythmia, 1907.7 RCs and Subject Orientation, 196

References, 204

8 Advanced Performance Architectures 207Aditya Singh, Aly Fathy, Isar Mostafanezhad, Jenshan Lin, Olga Boric-Lubecke,Shuhei Yamada, Victor M. Lubecke, and Yazhou Wang

8.1 DC Offset and Spectrum Folding, 2088.1.1 Single-Channel Homodyne System with Phase Tuning, 2088.1.2 Heterodyne System with Frequency Tuning, 2138.1.3 Low-IF Architecture, 220

8.2 Motion Interference Suppression, 2248.2.1 Interference Cancellation, 2268.2.2 Bistatic Radar: Sensor Nodes, 2318.2.3 Passive RF Tags, 240

8.3 Range Detection, 2508.3.1 Physiological Monitoring with FMCW Radar, 2508.3.2 Physiological Monitoring with UWB Radar, 251References, 266

9 Applications and Future Research 269Aditya Singh and Victor M. Lubecke

9.1 Commercial Development, 2699.1.1 Healthcare, 2699.1.2 Defense, 272

9.2 Recent Research Areas, 2729.2.1 Sleep Study, 2729.2.2 Range, 275

�

� �

�

CONTENTS ix

9.2.3 Multiple Subject Detection, 2769.2.4 Animal Monitoring, 279

9.3 Conclusion, 282References, 282

Index 285

�

� �

�

�

� �

�

LIST OF CONTRIBUTORS

Olga Boric-Lubecke, Department of Electrical Engineering, University of Hawaii atManoa, Honolulu, Hawaii, United States

Amy D. Droitcour, Wave 80 Biosciences, Inc., San Francisco, California, UnitedStates

Aly Fathy, Department of Electrical Engineering and Computer Science, Universityof Tennessee, Knoxville, Tennessee, United States

John Kiriazi, QCT RF Systems, Qualcomm Inc., San Diego, California, UnitedStates

Jenshan Lin, Department of Electrical and Computer Engineering, University ofFlorida, Gainesville, Florida, United States

Victor M. Lubecke, Department of Electrical Engineering, University of Hawaii atManoa, Honolulu, Hawaii, United States

Wansuree Massagram, Department of Computer Science and Information Technol-ogy, Naresuan University, Phitsanulok, Thailand

Isar Mostafanezhad, Department of Physics and Astronomy, University of Hawaiiat Manoa, Honolulu, Hawaii, United States

Byung-Kwon Park, DAS Sensor SW Engineering Team, Hyundai Mobis Mecha-tronics R&D Center, Gyeonggi-Do, South Korea

Aditya Singh, University of Hawaii Neuro-science and MRI research Program,John A. Burns School of Medicine, Honolulu, Hawaii, United States

�

� �

�

xii LIST OF CONTRIBUTORS

Alex Vergara, Theranova LLC, San Francisco, California, United States

Yazhou Wang, Boston Design Center, RF Micro Devices, Inc., Billerica,Massachusetts, United States

Shuhei Yamada, Department of Electrical Engineering, University of Hawaii atManoa, Honolulu, Hawaii, United States

Ehsan Yavari, Department of Electrical Engineering, University of Hawaii at Manoa,Honolulu, Hawaii, United States

�

� �

�

1INTRODUCTION

Amy D. Droitcour1, Olga Boric-Lubecke2, Shuhei Yamada2,and Victor M. Lubecke2

1Wave 80 Biosciences, Inc., San Francisco, California, United States2Department of Electrical Engineering, University of Hawaii at Manoa, Honolulu, Hawaii,United States

Noncontact detection and monitoring of human cardiopulmonary activity is one ofthe most promising solutions for sleep monitoring, postsurgery monitoring, homehealth care, and search and rescue applications. Without contact or subject prepa-ration (special clothing, attachments, etc.), this could facilitate health monitoring tothe chronically ill, enable sleep monitoring outside of sleep laboratories, detect sur-vivors under rubble, and deliver warnings of emergencies or changes in conditions ofpatients. Doppler radar remote sensing of physiological signatures has shown promiseto this end. The development of Doppler radar for remote sensing of vital signs,with proof of concept demonstrated for various applications [Li et al., 2013], couldoffer a platform for unobtrusive, noncontact, yet continuous physiological monitoringsystems.

Cardiopulmonary monitoring is typically carried out with contact sensors suchas electrocardiogram (ECG) electrodes. The use of contact sensors is neitherpossible nor desirable in many situations, due to, for example, skin irritation, orsimply lack of access for direct contact. The long-term, continuous use of contactsensors is also limited by degradation in contact quality over time. Some examplesof long-term health-care monitoring that would benefit from noncontact sensinginclude monitoring postsurgery patients, chronic and elderly patients, and patientswith sleep disorders. Premature infants and burn victims will also clearly benefit

Doppler Radar Physiological Sensing, First Edition.Edited by Olga Boric-Lubecke, Victor M. Lubecke, Amy D. Droitcour, Byung-Kwon Park, and Aditya Singh.© 2016 John Wiley & Sons, Inc. Published 2016 by John Wiley & Sons, Inc.

�

� �

�

2 INTRODUCTION

from noncontact sensing due to compromised skin integrity. A failure to respond topatient deterioration promptly and appropriately can lead to increased morbidity andmortality, increased requirement for intensive care, and elevated costs [Tarassenkoet al., 2006]. Early identification of patient deterioration is important, as it canprevent subsequent cardiopulmonary arrest and reduce mortality. Early recognitionof physiological abnormalities coupled with the rapid intervention of suitablytrained staff may result in an improvement in the functional outcome or mortalityrate. Early recognition relies on the physiological observations being measuredaccurately and at intervals appropriate to the condition of the patient. However,many patients are not monitored regularly, and some vital signs such as respiratoryrate are measured significantly less frequently than other vital signs. There is a needfor straightforward, automated, continuous physiological monitoring technology.

Noncontact physiological monitoring may make a significant impact beyondhealth-care applications, especially in situations where direct access to the subjectis not available. Such situations include, for example, occupancy sensors for energyefficiency [Yavari et al., 2014], search and rescue operations for survivor detectionunder rubble [Chuang et al., 1990], and detection of adversaries through walls[Lubecke et al., 2007].

1.1 CURRENT METHODS OF PHYSIOLOGICAL MONITORING

Assessment of cardiopulmonary functions is most often performed with contact sen-sors when direct access to the subject is available. ECG is the gold standard for heartmonitoring that is often used in hospital and ambulatory settings, whereas there isno equivalent gold standard for respiratory monitoring. There are many differentapproaches used for respiratory monitoring; however, none of them are easily applied.Even though respiratory rate is a key early indicator of physiological instability thatmay lead to a critical event, respiratory rate is measured significantly less frequentlythan other vital signs, such as blood pressure, pulse rate, and arterial oxygen satura-tion. Among the vital signs, respiratory rate is the only sign that is typically measuredmanually, via visual assessment, with a nurse counting chest excursions.

The current practices to measure respiration are divided into three categories:measurement of oxygen saturation, measurement of airflow, and measurement ofrespiratory effort/movement [Webster, 2010]. Pulse oximetry measures the percent-age of hemoglobin (Hb) that is saturated with oxygen. A source of light originatesfrom the probe at two wavelengths (650 and 805 nm). The light is partly absorbed byhemoglobin, at amounts which differ depending on whether it is saturated or desatu-rated with oxygen. Direct measurement of airflow typically uses a spirometer with amouth piece or a face mask. It contains a precision differential pressure transducer forthe measurements of respiration flow rates. The spirometer records the volume andrate of air that is breathed in and out over a specified time. These spirometers are rarelyused continuously because they have large dead volumes and high resistance, whichmake them unpleasant to use. Indirect measurement of airflow, such as with a ther-mocouple or capnograph, has less adverse effects, but still requires the placement of

�

� �

�

NEED FOR NONCONTACT PHYSIOLOGICAL MONITORING 3

sensors in front of the nose and/or mouth. Respiratory effort/movement measurementcan be monitored by measuring body volume changes; transthoracic inductance andimpedance plethysmographs, strain gauge measurement of thoracic circumference,pneumatic respiration transducers, and whole-body plethysmographs are examplesof indirect techniques. Each respiratory measurement method has unique advantagesand disadvantages. Pulse oximetry measurements indicate that a respiratory distur-bance has occurred, but do not provide respiratory rate. Airflow measurements are themost accurate, but interfere with normal respiration. The whole-body plethysmographcan be highly accurate and does not interfere with respiration, but requires immobi-lization of the patient. The performance of commonly used transducers (belts or elec-trodes) for ambulatory respiration monitoring significantly degrades over time withwear and tear. Impedance plethysmography, performed through ECG electrodes, isthe most common method of continuously measuring respiratory rate in the hospital.

The electrocardiograph (ECG) is traditionally considered the standard way to mea-sure the cardiac activity. It records the electrical activity of the heart over time. Elec-trical waves cause the heart muscle to contract. These waves pass through the bodyand can be measured at electrodes attached to the skin. Electrodes on different sidesof the heart measure the activity of different parts of the heart muscle. An ECG dis-plays the voltage between pairs of these electrodes, and the muscle activity that theymeasure, from different directions. This display indicates the overall rhythm of theheart, and weaknesses in different parts of the heart muscle. The other approach ispulse measurement of changes in blood volume in the skin. Pulse measurements, suchas a photoplethysmograph (PPG) or piezoresistance, use optical or pressure sensorsto identify pulses of blood driven by heartbeats. These are less invasive and simplerthan ECG, yet both of these methods require patients to be tethered to the sensingdevices.

1.2 NEED FOR NONCONTACT PHYSIOLOGICAL MONITORING

The ability to remotely detect vital signs such as heart beat and respiration is partic-ularly useful in situations where direct contact with the subject is either impossibleor unwanted. Avoidance of problems such as skin irritation, restriction of breathing,and electrode contacts is desirable in a number of health-care applications, includ-ing monitoring of patients with compromised skin, and sleep monitoring. Beyondhealth care, the potential applications that could benefit from remote sensing of phys-iological signals include fatigue monitoring, border crossing monitoring, occupancysensors, sense through the wall, and search and rescue operations.

1.2.1 Patients with Compromised Skin

Development of reliable noninvasive physiological monitoring is an important goalin modern health-care research. Knowledge of routinely monitored heart and respi-ratory patterns would be clinically useful in many situations. In neonatal intensivecare units, infants often suffer skin damage from adhesive tape, electrocardiogram

�

� �

�

4 INTRODUCTION

electrodes, electroencephalogram electrodes, and transcutaneous probes, with somelesions leaving scars [Colditz et al., 1999]. Monitoring the cardiac state of burn vic-tims can be challenging because it is sometimes difficult to find enough skin on whichto apply an ECG electrode. Sometimes the electrode is stapled to the skin, or to anundebrided burn area [Loo et al., 2004]. Often an esophageal ECG must be usedbecause adequate skin area cannot be located. A wireless heart and respiration ratemonitor could fill the needs for both neonates and burn victims, by enabling the mon-itoring of these vital signs without contacting the skin with electrodes.

1.2.2 Sleep Monitoring

Cardiopulmonary activity is the main parameter used in the study of sleep disorders.Sleep is widely understood to play a key role in physical and mental health. Thequality and quantity of sleep that an individual gets can have a significant impact onlearning and memory, metabolism and weight, safety, mood, cardiovascular health,disease, and immune system function. Obstructive sleep apnea (OSA) is the mostcommon sleep disorder with an estimated 12 million Americans suffering from it.Risk factors include gender, weight and age (being male, overweight, and over theage of 40), but sleep apnea can strike anyone at any age, even children. One of everyfive adults has at least mild OSA, and one of fifteen has at least moderate OSA [Younget al., 2002]. OSA has many negative effects, including excessive daytime sleepi-ness, increased risk of motor vehicle accidents, hypertension, psychological distress,and cognitive impairment. Apnea is the cessation of airflow for 10 s or longer, andOSA is apnea that occurs in spite of respiratory effort. To differentiate between cen-tral and obstructive apneic events, measurements of respiratory movement must bemade in addition to measurements of airflow [Phillips et al., 1998]. Current laboratorypolysomnography (PSG) is cumbersome, inconvenient, and expensive, causing con-siderable interest in portable monitoring of the condition. A Doppler radar monitoringsystem could identify respiratory movement, without the difficulties that accompanya full polysomnographic recording [Singh et al., 2013].

The gold standard for the clinical diagnosis of obstructive sleep apnea syndrome(OSAS) is PSG, consisting of simultaneous recordings of electrophysiological andrespiratory signals, and overnight monitoring of the patient in a specially equippedsleep laboratory [Kryger et al., 2000]. However, the scarcity of sleep clinics and theexpense associated with standard PSG allows treatment of small numbers of OSAScases. The lack of awareness among the public and health-care professionals resultsin the vast majority of sufferers remaining undiagnosed and untreated, despite the factthat this serious disorder can have significant consequences. Untreated sleep apneacan cause high blood pressure and other cardiovascular disease, memory problems,depression, weight gain, and headaches. Moreover, untreated sleep apnea may beresponsible for job impairment and motor vehicle collisions. A simple, less costly,noninvasive, reliable and ambulatory screening method for OSAS is desirable.

�

� �

�

DOPPLER RADAR POTENTIAL FOR PHYSIOLOGICAL MONITORING 5

Sudden infant death syndrome (SIDS) is believed to be attributable to sleep apnea.Although rates of SIDS have declined sharply in the past 15 years, SIDS is still thethird leading cause of infant mortality and the leading cause of postneonatal infantmortality [Arias et al., 2003; Hunt and Hauck, 2004]. In 2001, 8.1% of all infantdeaths were caused by SIDS, affecting 55.5 of every 100,000 live births. Appar-ent life-threatening events (ALTEs), defined as an episode that is characterized bysome combination of apnea, color change, marked change in muscle tone, chok-ing, or gagging, are experienced by 2.46 of every 1000 infants [Kiechl-Kohlendorferet al., 2005]. Although home electronic surveillance does not reduce the risk of SIDSat this time, this may be due to limits of current home-monitoring devices, or highfalse-negative rates. If obstructed breathing, central apnea, bradycardia, or oxygensaturation could be reliably detected, intervention could save infants’ lives [Hunt andHauck, 2004]. A Doppler radar device could detect central apneic events, where thereis no respiratory motion, and bradycardia, where the heart rate slows. Doppler radarcould be an integral part of a combination of sensors that could provide reliable homeSIDS monitoring.

1.2.3 Elderly Monitoring

The population share of the elderly around the globe has been steadily increasing,due to improvements in health care and decrease in birth rates. The global percentageof people aged 60 years or above increased from 9.2% in 1990 to 11.7% in 2013and will continue to grow as a proportion of the world population, projected to reach21.1% by 2050 [United Nations, 2013]. This has resulted in an increasing need forhealth-care equipment specialized for routine in home monitoring of the elderly. Withthe reduced mobility, elderly adults may be at high risk of gait or balance disorders,which are the major causes of fall in this population and risk factors for increasingmorbidity and mortality. Therefore, more specialized health-care equipment is neededfor long-term monitoring of gait for the elderly.

The Doppler radar has demonstrated potential for human gait monitoring [Wangand Fathy, 2011]. A Doppler radar-based approach for gait monitoring and fall detec-tion was proposed, with good accuracy in distinguishing common fall events fromnormal movements [Mercuri et al., 2013]. Such a system could be potentially linkedto medical monitoring personnel to provide a prompt alert in the event of emergencies.

1.3 DOPPLER RADAR POTENTIAL FOR PHYSIOLOGICALMONITORING

The development of Doppler radar for sensing of cardiopulmonary sensing, withproof of concept demonstrated for various applications [Li et al., 2013], could offera platform to establish noncontact, continuous, physiological monitoring systems.

�

� �

�

6 INTRODUCTION

Doppler radar systems can perform noncontact sensing of respiratory and cardiacsignatures at a distance, through clothing, walls, or debris. A particular advantage ofDoppler radar is its ability to detect both heart and respiratory signals simultaneously,but independently, and without contact with the subject. This may be particularlyuseful in chronic disease management, sleep studies, heart rate variability (HRV)and energy balance studies, and for obtaining biometric signatures for securityapplications.

1.3.1 Principle of Operation and Power Budget

Radar, an acronym for RAdio Detection And Ranging, describes a system thattransmits an electromagnetic signal and senses the echo from reflecting objects,thereby gaining information about those objects. The time delay between thetransmitted and received signals indicates the distance to the target; the frequencyshift of the received signal due to Doppler effect enables calculation of the target’svelocity; and the strength of the signal gives information about the target’s radarcross section, which provides information about its size, geometry, and composition.A major advantage of radio and microwave frequency radar systems is that thesewaves can penetrate through some objects that light cannot penetrate, allowingdetection of objects that cannot be seen. However, radar systems developed fordifferent applications may operate at many different frequencies, varying from a fewmegahertz to optical frequencies.

The Doppler effect can be observed as the change of frequency or pitch whena wave source moves either toward or away from the observer. This principle wasdiscovered by the Austrian physicist Christian Doppler in 1842, and it applies to allwave motion. Doppler radar uses this principle to measure target velocity from thefrequency shift between the transmitted electromagnetic wave and the wave reflectedfrom the moving object.

Radar systems were originally developed for military applications includingsurveillance and weapon control. Radar now has many civil applications, includingnavigation of aircraft, ships, and spacecraft, remote sensing of the environment(including weather), and law enforcement. Depending on the radar system hardwareand the type of signal sent, it may be possible to detect the range and/or angle to thetarget, the size and shape of the target, and the linear and/or rotational velocity of thetarget [Skolnik, 1990]. Depending on which of these parameters is most important tosense, as well as the range to and the nature of the target, different radar topologiesmay be used. A pure continuous-wave (CW) system can readily detect movingtargets via the Doppler shift of the received signal, although it cannot detect therange. Frequency-modulated continuous-wave (FMCW) radar systems can detectboth the range to and the velocity of the target. Altimeters and Doppler navigationdevices use FMCW radar systems [Saunders, 1990]. Pulsed radar allows transmittingand receiving to occur at different times, and it is used when the return signal is muchsmaller than the transmitted signal and, therefore, difficult to sense the receivedsignal in the presence of the transmitted signal [Skolnik, 1990]. Different types ofDoppler radar and their applications are discussed in more detail in Chapter 2.

�

� �

�

DOPPLER RADAR POTENTIAL FOR PHYSIOLOGICAL MONITORING 7

A low-power Doppler radar system can be used to sense physiological movement,enabling the monitoring of vital signs parameters without contact, and throughclothing and bedding [Li et al., 2013]. A low-power radio frequency (RF) signalis transmitted, and as it reflects from the patient’s body, the echo is modulated byphysiological motion. The Doppler shift theory states that a reflected signal froman object with periodic movement but zero net velocity is phase modulated. Thisphase is proportional to the displacement of the subject. If the subject is the humanbody, the reflected signal will contain information on the positional variations on thesurface due to cardiopulmonary activity. A combination of hardware and softwarecompares the echo signal with the transmitted signal, and extracts a physiologicalmotion signal.

CW, FMWC, and pulsed Doppler radar can be used to sense physiological move-ment. A CW radar topology is the simplest radar topology for two reasons: a singleoscillator can be used for both the transmitter and the receiver, and the extremely nar-row signal bandwidth avoids interference and rejects stationary clutter. A pure CWradar system can measure targets at any range (subject to the signal-to-noise ratio(SNR)) that are moving at any velocity (subject to the receiver bandwidth) withoutambiguity, unlike pulsed or modulated systems that have limited velocity resolution.However, CW radar systems cannot detect range without modulation, and when mod-ulated, the same range ambiguities found in pulsed radar systems are present.

When the goal of the measurement is target motion rather than distance to thetarget, a pure CW radar system is very effective. When the CW signal is directed ata target, it is reflected and frequency-modulated by the target motion. According toDoppler theory, a target with a time-varying position but no net velocity will reflectthe signal, modulating its phase in proportion to the time-varying position of the tar-get. A stationary person’s chest has a periodic movement with no net velocity, anda CW radar with the chest as the target will, therefore, receive a signal similar tothe transmitted signal, with its phase modulated by the time-varying chest position.Demodulating the phase will then provide a signal directly proportional to the chestposition, which contains information about movement due to heartbeat and respira-tion, from which heart and respiration rates can be determined. Noncontact heart andrespiration monitors have been developed based on this principle [Lin, 1992].

The most significant issues with CW radar are linked to its nature of constantlytransmitting and receiving, which results in the inability to separate reflections tem-porally. A portion of the transmitted signal leaks from the transmitter to the receiver,either through coupling between the transmit and receive circuitry, or directly throughthe antenna(s). In addition, clutter reflects some of the signal and its noise sidebandsback to the receiver, adding to the signal power at the transmit frequency due to leak-age. These unwanted signals result in a DC offset and low-frequency noise if they arenot eliminated before the signal is detected.

1.3.1.1 Frequency and Power Considerations The peak-to-peak chest motiondue to respiration in adults ranges from 4 to 12 mm [DeGroote et al., 1997; Kondoet al, 1997], while the peak-to-peak motion due to the heartbeat is about 0.5 mm[Ramachandran and Singh, 1989]. The amount of phase modulation due to chest

�

� �

�

8 INTRODUCTION

motion will be proportional to displacement and the operating frequency. At 2.4 GHz,1 cm of displacement corresponds to about 1 rad of phase change, and the phasechange increases proportionally to increasing frequency. The ability of the systemto discern changes in phase will depend on the overall SNR that is determined by thesize of the moving surface, displacement amplitude, range to target, and electricalproperties of the radar system.

The electrical properties of biological tissue affect the amount of signal that isreflected and transmitted, both at the skin–air interface and at interfaces between dif-ferent tissues within the body. The electric properties of biological tissue depend onfrequency of operation, and are largely governed by the percentage of water content.Tissues with high water content, such as skin, muscle, and blood, are more lossy andmore readily absorb electromagnetic waves. As the frequency of operation increases,the losses increase, whereas tissues with low water content, such as bone and fat, arelargely transparent to electromagnetic waves. The tissue contrast based on differentabsorption and propagation characteristics has been used for microwave imaging.

Doppler radar detects all motion in the radar field of view. If the antenna is placedin contact with the skin, internal organ motion may be measured, assuming that thereis enough penetration for the electromagnetic wave to reach the internal organs, andpropagate back to the receiver after the reflection. If the Doppler radar is placed atsome standoff distance from a subject, at frequencies of 2.4 GHz and above, about50% of the incident power will be reflected at the air–skin interface, and more than90% of the reflected power will come from the surface reflection. Thus, in this case,for all practical purposes, we can assume that noncontact Doppler radar physiologicalmeasurements are measurements of skin surface motion.

1.3.2 History of Doppler Radar in Physiological Monitoring

Physiological motion detection with CW Doppler radar has been known since the1970s [Lin, 1975], and with FMCW [Sharpe, 1990] and ultra-wideband (UWB)[Staderini, 2002] Doppler radar since the 1980s. Understanding of microwavenoninvasive physiological sensing has advanced tremendously in the past fewdecades, and recent advances in wireless technology have enabled further progressin medical radar, culminating with recent FDA approvals. Widespread use ofmicrowave technology and digital processors in common household communica-tions devices has driven down costs, making it possible to develop practical radarmonitors that cost significantly less than conventional cardiopulmonary assessmentinstruments.

Microwave Doppler radar monitoring of respiratory and cardiac movements wasfirst demonstrated in the late 1970s, when respiration [Lin, 1975; Lin et al., 1977] andheart beats [Lin et al., 1979] were measured separately, with a breath-hold requiredfor the heart measurement [Lin et al., 1979]. X-band sweep oscillators were used,with horn antennas directing the microwave energy toward the upper torso of the sub-jects. A nonanesthetized rabbit’s respiration was measured from a distance of 30 cm[Lin, 1975]. In [Lin et al., 1977], the same system was used with an apnea-detector

�

� �

�

DOPPLER RADAR POTENTIAL FOR PHYSIOLOGICAL MONITORING 9

circuit, and was tested on a rabbit and two cats, all of which were anesthetized andintubated. Both hyperventilation and apnea were induced in the animals, and bothstates were clearly detected by the microwave monitor. Microwave apexcardiographywas demonstrated by using a continuous-wave 2-GHz microwave signal with anantenna placed 3 cm above the apex, and the precordial motions were easily detected[Lin et al., 1979].

From the mid-1980s through the late 1990s, radar transceivers were developed thatincorporated analog and digital signal processing to separate the small heart signalfrom the much larger respiration signal, so the subject did not need to hold his/herbreath for the heart rate to be measured, and heart and respiration could be mea-sured simultaneously [Chan and Lin, 1987; Chen et al., 2000; Chuang et al., 1990;Greneker, 1997; Seals et al., 1986]. These transceivers were used for the detection ofheart and respiration rates of persons in rubble, persons behind walls, and Olympicathletes. An analog amplification and filtering for separation of heart and respirationsignatures was combined with 8-bit digitization and digital signal processing to detectheart and respiration rates [Chan and Lin, 1987]. An automatic clutter-cancellationcircuit was developed to facilitate measurement of the heart and respiration signa-tures through seven layers of brick [Chuang et al., 1990] and through 10 ft of rubble[Chen et al., 2000]. Heart and respiration rates of athletes were successfully detectedat ranges exceeding 10 m [Greneker, 1997]. At 100 m standoff distance, the limit wasmoving background clutter, not the system sensitivity. A quadrature receiver was usedto avoid phase-demodulation null points [Seals et al., 1986].

More recently, Doppler radar vital signs monitoring was explored to detect hypov-olemic states and shock in persons under rubble or in biochemical hazard conditionsthat could pose danger to health-care providers [Matsui et al., 2004a, 2004b]. Hypo-volemic rabbits and rabbits in shock could be reliably distinguished from the controlrabbits based on the Doppler radar information by using linear discriminant analy-sis on the heart and respiration rates. The hypovolemic rabbits have higher heart andrespiration rates. Doppler radar was also used to estimate arterial blood pH withoutcontact using heart and respiration rate monitoring coupled with an infrared thermo-graphic temperature measurement and an exhaled gas (CO and CO2) analyzer [Matsuiet al., 2006]. This measurement was successful at estimating blood pH using linearregression analysis on hypovolemic rabbits.

The connection of this technology to the existing wireless communications infras-tructure was also investigated [Boric-Lubecke and Lubecke, 2002; Lubecke et al.,2002; Boric-Lubecke et al., 2003]. A modified wireless LAN PCMCIA card wasused to detect heart and respiration [Boric-Lubecke et al., 2003], and a module thatcombines the transmitted and reflected signals from any wireless communicationdevice, such as a cordless telephone, was used to detect heart and respiration [Lubeckeet al., 2002]. Using this technology to directly connect Doppler measurement of heartand respiration rate to health-care providers has been proposed [Boric-Lubecke andLubecke, 2002].

In addition, UWB radar has been used for the measurement of heart and respi-ration rates. Using 0.4-W pulses and a 1-GHz central frequency, heart rates were

�

� �

�

10 INTRODUCTION

detected through 1 m of air and a 0.4-m brick wall [Immoreev and Samkov, 2005]and respiration was measured at up to 5 m [Ossberger et al., 2004].

Research efforts in the last decade have been moving the technology developmenttoward lower power, lighter weight, smaller form factor, better accuracy, longerdetection range, and more robust operation for practical portable and handheldapplications. Among many possible applications this technology can be used for,health care seems to be drawing the most interest. As an example, a baby monitorusing this technology was demonstrated to monitor SIDS [Hafner et al., 2007; Liet al., 2009]. The Doppler radar embedded into the baby monitor detects tiny babymovements induced by breathing. If no movement is detected within 20 s, an alarmgoes off to warn the parents. Operating in similar ways, biomedical Doppler radar isalso being investigated as a cost-effective solution for long-term monitoring of sleepapnea [Singh et al., 2013]. Human studies in clinical environment have validated thistechnology as a potential substitute for conventional respiratory monitors [Droitcouret al., 2009; Massagram et al., 2009] and a useful tool for precise assessment of keyparameters relating to cardiopulmonary activity including body orientation [Kiriaziet al., 2012]. Furthermore, recent studies have demonstrated that Doppler radarcould help a medical linear accelerator to track the location of a mobile tumor duringradiotherapy with the help of advanced signal-processing algorithms [Li et al., 2011;Gu et al., 2012]. Doppler radar has also been applied to monitor the health andbehavior of land and sea animals including lizards and fish [Singh et al., 2012b,2012a; Hafner et al., 2012].

With growing interests in health and life sciences in the engineering commu-nity, many researchers have been contributing to the technology advancement inthis field. This has led to various radar front-end architectures including the con-ventional homodyne/heterodyne [Xiao et al., 2006], self-/mutual-injection locking[Wang et al., 2011], and coherent low-IF [Mostafanezhad and Boric-Lubecke, 2014].Each of these architectures shows its specific advantage in certain environments. Sig-nal conditioning and processing methods such as adaptive DC calibration [Vergaraet al., 2008b; Gu and Li, 2012], arctangent demodulation [Park et al., 2007], andnoise cancellation [Li and Lin, 2008; Oum et al., 2008; Fletcher and Jing, 2009;Wiesner, 2009] have been proposed to enable practical applications of the biomedicalradar. Various techniques, including blind source separation (BSS) and the use of pas-sive RF tags, have been applied to isolate multiple targets and clutter noise [Vergaraet al., 2008a; Singh and Lubecke, 2013]. Hardware implementations from benchtopfast prototyping using fundamental RF/microwave instruments [Gu et al., 2010] toradar-on-chip application-specific integrated circuits (ASICs) [Droitcour et al., 2002,2004; Li et al., 2008, 2010] have been demonstrated to make this technology availableto both researchers and general users.

The selection of published works outlining Doppler radar physiological moni-toring history from 1975 to date are described briefly in Table 1.1 with the year ofpublication, reference, description, and results.

�

� �

�

TA

BL

E1.

1D

oppl

erR

adar

Phy

siol

ogic

alM

onit

orin

gfr

om19

75to

2014

Ref

eren

ces

Des

crip

tion

Res

ults

Lin

[197

5]X

-ban

dsw

eep

osci

llato

r,re

ctan

gula

rho

rnan

tenn

aR

espi

ratio

nm

easu

red

onra

bbit

and

hum

anat

30-c

mra

nge

Lin

etal

.[19

77]

Mic

row

ave

apne

ade

tect

orpr

opos

edfo

rlo

w-b

irth

-wei

ght

infa

nts

3-m

Wat

30-c

m,3

-cm

by3-

cmho

rnR

espi

ratio

nm

easu

red

onca

tsan

dra

bbit

at30

-cm

rang

eD

etec

tion

ofap

nea

and

hype

rven

tilat

ion

Lin

etal

.[19

79]

“Ape

xcar

diog

raph

y”he

artm

easu

rem

ents

mad

ew

hile

brea

thhe

ldSi

gnal

ampl

itude

and

phas

eva

ryw

ithan

tenn

alo

catio

n

Hea

rtm

easu

rem

entm

ade

3-cm

from

apex

Seal

set

al.[

1986

]A

nalo

gsi

gnal

proc

essi

ngse

para

tes

hear

tand

resp

irat

ion

sign

sD

igita

lrat

ede

tect

ion

algo

rith

m10

-an

d3-

GH

zqu

adra

ture

rada

rsy

stem

s

Mea

sure

dhe

arta

ndre

spir

atio

nsi

mul

tane

ousl

yw

ithdi

gita

lrat

ede

tect

ion

algo

rith

m

Che

net

al.[

2000

]A

nX

-ban

dm

icro

wav

elif

e-de

tect

ion

syst

emha

sbe

ende

velo

ped

for

vita

lsig

nsM

easu

red

hear

tbea

tand

brea

thin

gof

hum

ansu

bjec

tsly

ing

onth

egr

ound

ata

dist

ance

ofab

out3

0m

orlo

cate

dbe

hind

aci

nder

bloc

kw

all

Cha

nan

dL

in[1

987]

Hea

rtan

dre

spir

atio

nse

para

ted

with

anal

ogan

ddi

gita

lsi

gnal

proc

essi

ng:a

mpl

ifica

tion,

filte

ring

,8-b

itA

DC

sam

pled

at80

Hz

10.5

-GH

z,10

-mW

,hor

nan

tenn

aa

few

cent

imet

erfr

omsu

bjec

t

Hea

rtan

dre

spir

atio

nob

tain

edsi

mul

tane

ousl

yat

5–7-

cmra

nge

Chu

ang

etal

.[19

90]

Hea

rtan

dre

spir

atio

nfo

rde

tect

ing

vict

ims

incl

utte

r10

-GH

zsy

stem

does

notp

enet

rate

wet

bric

ks,b

ut2-

GH

zsy

stem

does

Aut

omat

iccl

utte

rca

ncel

latio

nal

gori

thm

intr

oduc

ed

Succ

essf

ulm

easu

rem

entw

ithsu

bjec

tbot

hfa

ceup

and

face

dow

n,th

roug

h4–

7la

yers

ofbr

ick,

with

2-an

d10

-GH

zra

dar

syst

ems

(Con

tinu

ed)

11

�

� �

�

TA

BL

E1.

1(C

ontin

ued)

Ref

eren

ces

Des

crip

tion

Res

ults

Gre

neke

r[1

997]

0.6-

mdi

shai

med

atsu

bjec

t’sth

orax

24-G

Hz,

30-m

Wou

tput

sign

al,4

0-dB

ante

nna

gain

Hea

rtbe

atan

dre

spir

atio

nm

easu

red

atra

nges

exce

edin

g10

mC

hen

etal

.[20

00]

Lif

e-de

tect

ion

syst

emfo

rvi

ctim

sun

der

rubb

leor

behi

ndba

rrie

rsin

clud

ing

6in

.of

stee

l,br

icks

,and

cylin

der

bloc

ksH

eart

and

resp

irat

ion

wer

em

easu

red

450-

MH

zsi

gnal

pene

trat

esde

epes

tint

oco

ncre

teru

bble

with

outm

etal

1150

-MH

zsi

gnal

pene

trat

esru

bble

with

met

allic

wir

em

esh

Dro

itcou

ret

al.[

2002

]T

hefir

stC

MO

San

dB

iCM

OS

Dop

pler

rada

rch

ips

Afu

llyin

tegr

ated

dire

ctco

nver

sion

Dop

pler

rada

rth

atde

tect

she

arta

ndre

spir

atio

nm

ovem

enta

tadi

stan

ceof

50cm

.The

1.6

GH

ztr

ansc

eive

ris

impl

emen

ted

inbo

thC

MO

San

dB

iCM

OS

tech

nolo

gies

Lub

ecke

etal

.[20

02]

Add

-on

mod

ule

uses

sign

als

from

exis

ting

wir

eles

sde

vice

sto

mea

sure

hear

tand

resp

irat

ion

rate

sH

eart

and

resp

irat

ion

wer

em

easu

red

with

a2.

4-G

Hz

cord

less

phon

ean

da

2.4-

GH

zsi

gnal

gene

rato

rB

oric

-Lub

ecke

etal

.[2

003]

Mod

ified

wir

eles

sL

AN

PCM

CIA

card

sar

eus

edto

sens

ehe

arta

ndre

spir

atio

nra

tes

Hea

rtan

dre

spir

atio

nw

ere

obta

ined

succ

essf

ully

at40

-cm

Dro

itcou

ret

al.[

2004

]R

ange

corr

elat

ions

and

I/Q

perf

orm

ance

bene

fits

insi

ngle

-chi

pD

oppl

erra

dars

The

rang

e-co

rrel

atio

nef

fect

onre

sidu

alph

ase

nois

eis

acr

itica

lfac

tor

whe

nde

tect

ing

smal

lpha

seflu

ctua

tions

with

ahi

gh-p

hase

-noi

seon

-chi

pos

cilla

tor.

Phas

e-no

ise

redu

ctio

ndu

eto

rang

eco

rrel

atio

nw

asex

peri

men

tally

eval

uate

dM

atsu

ieta

l.[2

004a

]D

eter

min

ing

hypo

vole

mic

and

shoc

kst

ates

usin

glin

ear

disc

rim

inan

tana

lysi

sH

eart

and

resp

irat

ion

rate

sof

hypo

vole

mic

rabb

its12

15M

Hz,

70-m

Wou

tput

pow

er

Lin

ear

disc

rim

inan

tana

lysi

sef

fect

ivel

ypr

edic

ted

hypo

vole

mic

stat

eof

10ra

bbits

Oss

berg

eret

al.[

2004

]U

ltra-

wid

eban

dpu

lse

rada

rW

avel

etsi

gnal

proc

essi

ngR

espi

ratio

nm

easu

red

at1–

5-m

and

thro

ugh

aw

alla

t85

-cm

12

�

� �

�

Imm

oree

van

dSa

mko

v[2

005]

Ultr

a-w

ideb

and

rada

r,1-

GH

zce

ntra

lfre

quen

cyM

easu

red

resp

irat

ion

thro

ugh

1-m

air

and

0.4-

mbr

ick

wal

lon

one

subj

ect

Mat

suie

tal.

[200

6]U

sed

1215

-MH

zra

dar,

infr

ared

ther

mog

raph

y,an

dex

hale

dC

O/C

O2

anal

yzer

toes

timat

ebl

ood

pHSi

gnifi

cant

corr

elat

ion

ofm

easu

red

and

calc

ulat

edbl

ood

pHon

rabb

itsX

iao

etal

.[20

06]

Het

erod

yne

Ka-

band

rada

rD

oubl

e-si

deba

ndtr

ansm

issi

onis

used

for

avoi

ding

null

dete

ctio

npo

int

Park

etal

.[20

07]

Use

d2.

4-G

Hz

quad

ratu

rere

ceiv

er.A

rcta

ngen

tde

mod

ulat

ion

with

DC

offs

etco

mpe

nsat

ion

tode

tect

hear

tand

resp

irat

ion

rate

Rob

usta

ndac

cura

teou

tput

data

obta

ined

from

arct

ange

ntde

mod

ulat

ion.

Unw

ante

dD

Cof

fset

was

succ

essf

ully

elim

inat

edH

afne

ret

al.[

2007

]V

itals

igns

sens

ing

usin

ga

baby

mon

itor

and

apa

ssiv

ese

nsor

node

Sign

alfr

oma

baby

mon

itor,

with

anad

ditio

nof

apa

ssiv

ese

nsor

isus

edto

dete

ctvi

tals

igns

Lia

ndL

in[2

008]

4–7-

GH

zqu

adra

ture

tran

scei

ver

with

the

com

plex

and

arct

ange

ntde

mod

ulat

ion

for

rand

omm

ovem

ent

canc

ella

tion

The

com

plex

sign

alde

mod

ulat

ion

ism

ore

favo

rabl

efo

rra

ndom

body

mov

emen

tcan

cella

tion.

The

arct

ange

ntde

mod

ulat

ion

has

the

adva

ntag

eof

elim

inat

ing

the

harm

onic

and

inte

rmod

ulat

ion

inte

rfer

ence

athi

ghfr

eque

ncie

sus

ing

high

gain

ante

nnas

Lie

tal.

[200

8]A

5G

Hz

doub

le-s

ideb

and

rada

rse

nsor

chip

The

diff

eren

tiala

rchi

tect

ure

has

the

adva

ntag

eof

redu

cing

loca

losc

illat

ion

leak

age.

The

5G

Hz

doub

le-s

ideb

and

syst

emca

nav

oid

null

dete

ctio

npo

int

byfr

eque

ncy

tuni

ngV

erga

raet

al.[

2008

b]D

Cof

fset

canc

ella

tion

and

DC

info

rmat

ion

pres

erva

tion

Dig

itally

cont

rolle

dvo

ltage

feed

back

and

cent

erfin

ding

pres

erve

sth

eim

port

antD

Cin

form

atio

nfo

rop

timal

extr

actio

nof

phas

ein

form

atio

nin

the

quad

ratu

resy

stem

Ver

gara

etal

.[20

08a]

Blin

dso

urce

sepa

ratio

nof

hum

anm

otio

nT

hesu

cces

sful

sepa

ratio

nof

card

iopu

lmon

ary

mot

ion

and

hand

mot

ion

for

asi

ngle

subj

ect

(Con

tinu

ed)

13

�

� �

�

TA

BL

E1.

1(C

ontin

ued)

Ref

eren

ces

Des

crip

tion

Res

ults

Wie

sner

[200

9]C

Wvi

tals

ign

rada

rw

ithen

hanc

edta

rget

dete

ctio

npr

obab

ility

ina

high

lycl

utte

red

envi

ronm

entw

ithra

nge

and

angl

e-of

-arr

ival

estim

atio

n

Are

spir

atio

nsi

gnat

ure

arou

nd0.

2H

zan

dhi

gher

harm

onic

sal

ong

with

hear

tbea

tsig

natu

rear

ound

1.3

Hz

are

clea

rly

disc

erni

ble

Mas

sagr

amet

al.

[200

9]2.

4-G

Hz

quad

ratu

resy

stem

with

linea

rde

mod

ulat

ion

met

hod

tode

tect

hear

trat

eva

riab

ility

(HR

V)

and

resp

irat

ory

sinu

sar

rhyt

hmia

(RSA

)

The

data

wer

eob

tain

edfr

om12

hum

ansu

bjec

tsin

seat

edan

dsu

pine

posi

tions

.Hig

hac

cura

cyin

extr

actin

gth

eH

RV

and

RSA

indi

ces

was

achi

eved

Dro

itcou

ret

al.[

2009

]T

hefir

stcl

inic

alva

lidat

ion

ofth

ere

spir

ator

yra

teac

cura

cyA

hum

anst

udy

ina

clin

ical

envi

ronm

entv

alid

ated

Dop

pler

rada

rte

chno

logy

asa

pote

ntia

lsub

stitu

tefo

rco

nven

tiona

lres

pira

tory

mon

itors

Lie

tal.

[200

9]In

fant

vita

lsig

nm

onito

rT

heD

oppl

erra

dar

embe

dded

into

the

baby

mon

itor

dete

cts

tiny

baby

mov

emen

tsin

duce

dby

brea

thin

g.If

nom

ovem

enti

sde

tect

edw

ithin

20s,

anal

arm

goes

off

tow

arn

the

pare

nts

Flet

cher

and

Jing

,200

9D

ual-

ante

nna

diff

eren

tialr

adar

fron

t-en

dsy

stem

for

shor

t-ra

nge

hear

trat

ean

dba

ckgr

ound

mot

ion

nois

ere

mov

al

Des

igne

dtw

ohe

lical

dire

ctio

nala

nten

nas

at2.

46-G

Hz

and

2.51

0-G

Hz

tode

tect

hear

trat

eat

0.5-

mdi

stan

cew

ithm

otio

nno

ise

canc

ella

tion

Lie

tal.

[201

0]A

dire

ct-c

onve

rsio

n5.

8-G

Hz

rada

rse

nsor

chip

with

1-G

Hz

band

wid

thA

naly

ses

onse

nsiti

vity

and

link

budg

etgu

ide

the

desi

gnof

high

-sen

sitiv

ityno

ncon

tact

vita

lsig

nde

tect

or.T

his

rada

rse

nsor

chip

isso

ftw

are

confi

gura

ble

tose

tthe

oper

atio

npo

inta

ndde

tect

ion

rang

efo

rop

timal

perf

orm

ance

Gu

etal

.[20

10]

Fast

prot

otyp

ing

usin

gfu

ndam

enta

lRF/

mic

row

ave

inst

rum

ents

Het

erod

yne

digi

talq

uadr

atur

ede

mod

ulat

ion

arch

itect

ure

that

help

sm

itiga

tequ

adra

ture

chan

neli

mba

lanc

ean

del

imin

ate

the

com

plic

ated

DC

offs

etca

libra

tion

requ

ired

for

arct

ange

ntde

mod

ulat

ion

14

�

� �

�

Wan

get

al.[

2011

]Si

ngle

ante

nna

self

-inj

ectio

n-lo

cked

(SIL

)ra

dar

pres

ente

dto

dete

ctvi

tals

igns

with

rand

ombo

dym

otio

nca

ncel

latio

n

Two-

rada

rar

ray

plac

ed2-

maw

ayat

2.4

GH

zis

impl

emen

ted

with

rand

ombo

dym

otio

nca

ncel

latio

nfo

rm

onito

ring

subj

ectj

oggi

ngon

trea

dmill

.Pro

vide

dth

eore

tical

basi

sfo

rde

term

inin

gsi

gnal

-to-

nois

esp

ectr

alde

nsity

ratio

Gu

etal

.[20

12]

DC

-cou

pled

adap

tive

tuni

ngra

dar

sens

orto

prec

isel

ym

easu

rere

spir

ator

ym

ovem

enti

nm

otio

n-ad

aptiv

era

diot

hera

py

Subm

illim

eter

accu

racy

was

achi

eved

inm

easu

rem

ent

for

resp

irat

ory

gatin

gan

dha

spo

tent

ialt

umor

trac

king

Kir

iazi

etal

.[20

12]

Dua

l-fr

eque

ncy

(2.4

and

5.8-

GH

z)ra

dar

for

card

iopu

lmon

ary

effe

ctiv

era

dar

cros

sse

ctio

n(E

RC

S)an

ddi

spla

cem

enti

nth

edi

rect

ion

ofin

cide

nce

Res

ults

show

edth

atan

ER

CS

isla

rger

for

the

back

ofth

eto

rso

and

smal

ler

for

the

side

than

toth

efr

ont,

whi

leth

ere

spir

atio

nde

pth

issm

alle

rin

the

pron

epo

sitio

nth

anin

supi

neSi

ngh

etal

.[20

12a]

Liz

ard

activ

itym

onito

ring

and

mot

ion

clas

sific

atio

nL

izar

dm

onito

ring

ina

nonc

onfin

edla

bora

tory

envi

ronm

enta

ndus

esm

ultip

leD

oppl

erra

dars

oper

atin

gat

10.5

25G

Hz

Haf

ner

etal

.[20

12]

Fish

hear

trat

em

onito

ring

The

hear

trat

ede

tect

edby

the

body

cont

actr

adar

mat

ched

the

hear

trat

ede

tect

edby

EC

GSi

ngh

etal

.[20

13]

Dop

pler

rada

rfo

rcl

inic

alsl

eep

mon

itori

ngD

oppl

erra

dar

isbe

ing

inve

stig

ated

asa

cost

-eff

ectiv

eso

lutio

nfo

rlo

ng-t

erm

mon

itori

ngof

slee

pap

nea

Sing

han

dL

ubec

ke[2

013]

2.45

-GH

zsy

stem

with

afr

eque

ncy

doub

ling

pass

ive

plan

arha

rmon

icta

gT

heha

rmon

icra

dar

syst

emis

olat

esth

ere

spir

atio

nof

ata

gged

hum

ansu

bjec

tfro

man

unta

gged

larg

esc

atte

ring

obje

ctM

osta

fane

zhad

and

Bor

ic-L

ubec

ke[2

014]

Coh

eren

tlow

-IF

rece

iver

arch

itect

ure

Mea

sure

men

tson

am

echa

nica

ltar

geta

nda

hum

ansu

bjec

tdem

onst

rate

asi

gnal

-to-

nois

era

tioim

prov

emen

tof

7dB

,whi

chca

nin

crea

seth

era

nge

ofop

erat

ion

by50

%

15

�

� �

�

16 INTRODUCTION

REFERENCES

Arias E, MacDorman MF, Strobino DM, Guyer B. Annual summary of vital statistics – 2002.Pediatrics 2003;112(6):1215–1230.

Boric-Lubecke O, Atwater G, Lubecke VM. Wireless LAN PC card sensing of vital signs.Proceedings of IEEE Topical Conference on Wireless Communications Technology; 2003.p 206–207.

Boric-Lubecke O, Lubecke VM. Wireless house calls: using communications technology forhealth care monitoring. IEEE Microwave Mag 2002;3(3):43–48.

Chan KH, Lin JC. Microprocessor-based cardiopulmonary rate monitor. Med Biol Eng Comput1987;25(9):41–44.

Chen K-M, Huang Y, Zhang J, Norman A. Microwave life-detection systems for search-ing human subjects under earthquake rubble or behind barrier. IEEE Trans Biomed Eng2000;47(1):105–114.

Chuang H-R, Chen Y-F, and Chen K-M. Microprocessor-controlled automaticclutter-cancellation circuits for microwave systems to sense physiological move-ments remotely through the rubble. Proceedings of the Instrumentation and MeasurementTechnology Conference; 1990. p 177–181.

Colditz PB, Dunster KR, Joy GJ, Robertson IM. Anetoderma of prematurity in associationwith electrocardiographic electrodes. J Am Acad Dermatol 1999;41(3):479–481.

DeGroote A, Wantier M, Cheron G, Estenne M, Pavia M. Chest wall motion during tidal breath-ing. J Appl Physiol 1997;83(5):1531–1537.

Droitcour AD, Boric-Lubecke, O, Lubecke VM, Lin, J. 0.25 μm CMOS and BiCMOSsingle-chip direct-conversion Doppler radars for remote sensing of vital signs. IEEE Inter-national Solid-State Circuits Conference 2002, ISSCC 2002. Digest of Technical Papers;2002. Vol. 1, p 348–349.

Droitcour AD, Boric-Lubecke O, Lubecke VM, Lin J, Kovacs GTA. Range correlation and I/Qperformance benefits in single-chip silicon Doppler radars for noncontact cardiopulmonarymonitoring. IEEE Trans Microwave Theory Tech 2004;52:838–848.

Droitcour AD, Seto TB, Byung-Kwon P, Yamada S, Vergara A, El Hourani C, Shing T, et al.,Non-contact respiratory rate measurement validation for hospitalized patients. Engineeringin Medicine and Biology Society, 2009, EMBC 2009. Annual International Conference ofthe IEEE; 2009. p 4812–4815.

Fletcher R, Jing H. Low-cost differential front-end for Doppler radar vital sign monitor-ing. IEEE MTT-S International Microwave Symposium Digest, 2009, MTT’09; 2009. p1325–1328.

Greneker EF. Radar sensing of heartbeat and respiration at a distance with applications of thetechnology. Radar 97 Conference Proceedings; 1997. p 150–154.

Gu C, Li C. DC coupled CW radar sensor using fine-tuning adaptive feedback loop. ElectronLett 2012;48:344–345.

Gu C, Li C, Lin J, Long J, Huangfu J, Ran L. Instrument-based noncontact Doppler radar vitalsign detection system using heterodyne digital quadrature demodulation architecture. IEEETrans Instrum Meas 2010;59:1580–1588.

Gu C, Li R, Zhang H, Fung A, Torres C, Jiang S, Li C. Accurate respiration measurement usingDC-coupled continuous-wave radar sensor for motion-adaptive cancer radiotherapy. IEEETrans Biomed Eng 2012;59:3117–3123.