10

IEEE EMBEDDED SYSTEMS LETTERS, VOL. 8, NO. 1, MARCH 2016

A Compact Portable Microwave Life-Detection

Device for Finding Survivors
F. JalaliBidgoli, S. Moghadami, Student Member, IEEE, and S. Ardalan, Senior Member, IEEE

AbstractIn this letter, an ultra-sensitive compact portable microwave life-detection device is introduced and implemented with
promising results. By utilizing Doppler effect-based systems, vital
signs such as heartbeats and breathing can be detected and can be
used for nding survivors under earthquake rubble, injured soldiers on battleelds and as lie detection device. This device is tested
in both simulated and realistic situations, and it can accurately detect crucial signs of life through highly dense construction materials of about 1.5 m thick and standard density materials of about
10 m while operating at 1.15 GHz center frequency.
Index TermsBiomedicine, breathing signal, clutter cancelation, detector, heartbeat signal, human, life detection, microwave,
transceiver, vital signs, wireless.

I. INTRODUCTION

RADITIONAL approaches such as utilizing dogs for

nding survivors under earthquake sediment or injured
soldiers on battleelds where seconds determine life or the
death are not sufcient due to time consumption and inaccuracy. Other blind searching methods are restricted to specic
construction materials. These limitations caught the attention of
RF engineers to build efcient and sensitive electronic systems
as an alternative to standard methods [1][16].
The principle behind the life detection systems is Doppler
effect [11]. At certain frequency, electromagnetic waves can
propagate through materials and objects with low attenuation.
If a wavefront at a specic frequency irradiates at a mass of
construction materials covering a human subject, it can penetrate the mass and reect a wave modulated by the movement of the subject. One key point of the proposed system is
efcient elimination of clutter from environmental and other
noise sources. After eliminating clutter, vital signs such as heartbeat and breathing can be decoded from the subjects modulated wave reections. Another key point is selecting an appropriate frequency for wavefront propagation. For instance, to nd
injured soldiers L-band frequencies are not suitable, requiring
high frequency bands [3][4]. Conversely, X-band frequencies
are not appropriate for highly dense concrete; so lower bands
should be utilized instead [6]. The main goal of this letter is
nding trapped humans under denser construction materials/

Manuscript received June 22, 2015; accepted September 27, 2015. Date of
publication October 09, 2015; date of current version February 25, 2016. This
manuscript was recommended for publication by M. Balakrishnan.
The authors are with the Centre for Analog and Mixed Signal (AMS), Department of Electrical Engineering, San Jose State University, CA 95192 USA
(e-mail: fjalali@ieee.org; siavash.moghadami@sjsu.edu; ardalan@ieee.org).
Color versions of one or more of the gures in this letter are available online
at http://ieeexplore.ieee.org.
Digital Object Identier 10.1109/LES.2015.2489209

Therefore, the proposed system operates at 1.15 GHz (L-band)

using two similar horn antennas as probes.
The complete schematic of the implemented system is shown
in Fig. 1. The system is tested in both realistic and simulated
environments. The simulated environment consists of multiple
layers of tightly packed bricks and concrete slabs lled with
sand and covered by a metallic wire mesh. Measurement results show that the thickness of rubble could reach up to 1.5 m
and 10 m in simulated and realistic situations respectively. The
proposed system has 4 major parts: high-frequency solid-state
circuits (Block 1 to 4 in Fig. 1), two horn antennas to irradiate
and receive reected waves, a microprocessor-based system for
eliminating clutter, and a computer to provide digital control
signals and monitor reected waves. This systems form factor
and attached battery makes it a good portable device that can be
used in nding survivors in various situations.
II. SYSTEM AND CIRCUITS DESCRIPTION
Fig. 1 shows the complete schematic of the proposed system.
Since there is a difference of 0.1 to 50 Hz between radiated and
reected waves, an extremely low phase noise radiated wave is
needed. There are two main solutions: use phase-locked oscillators (PLL-based) or utilize comb generator multipliers. This
letter utilizes the second approach due to better phase noise
compared with the PLO.
In Block 1, comb generator multiplier consists a signal amplier that amplies the output signal of an oven-controlled crystal
oscillator (OCXO) with 115 MHz center frequency (OX-175
from Vectron) and a step recovery diode (SRD). As shown in
Fig. 2, diode is driven by OCXO. Simulated behavior of SRD in
Fig. 3(a) shows peaks that can be considered as high frequency
short pulses and SRD voltage spectrum used in proposed life
detection system is shown in Fig. 3(b). In the spectrum 10th
harmonic lie at 1150 MHz frequency that is desired and can be
selected by utilizing appropriate lter at SRD output. The measured phase noises of 115 MHz (reference signal) and 1.15 GHz
(output signal) are demonstrated in Fig. 4. The measurement results show that for 1 and 10 Hz offset, phase noise of OCXO is
dBc Hz and
dBc Hz respectively. A two-stage amplier (Gali5 by Mini Circuits and AH102 from Agilent) is then
utilized to boost the output power to 26.5 dBm. In order to eliminate other harmonics produced by the SRD, a microstrip-based
interdigital lter has been used with 70 MHz bandwidth. As a
result attenuation is 22 dB at 1035 MHz and 23 dB at 1265 MHz.
The second block consists of two power dividers (balanced
and unbalanced), a detector and an RF LNA. Both balanced
and unbalanced power dividers use the Wilkinson topology. The
RF detector used in this block has dynamic range of 70 dB
with very high accuracy (
dB) and bandwidth from 0.1 to

1943-0663 2015 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

JALALIBIDGOLI et al.: A COMPACT PORTABLE MICROWAVE LIFE-DETECTION DEVICE FOR FINDING SURVIVORS

11

Fig. 1. Schematic of the proposed system.

Fig. 2. Applying the output of OCXO to SRD along with two-stage amplier.
Fig. 5. (a) Proposed band-pass lter in the receive path. (b) Measured insertion
loss of the proposed BPF.

Fig. 3. (a) Voltage of SRD pulses containing high frequency components.

(b) Output spectrum of SRD voltage in the proposed life detection system.

Fig. 4. Measured phase noises of output and reference signals.

2.5 GHz. Since the dynamic range of the utilized detector is

around 70 dB, subsequent ampliers can be implemented for

further amplication. The RF LNA used in this block consists

of two subsequent Gali5 and Gali74 ampliers and overall block
has 3 dB noise gure and at least 35 dB power gain. In order to
achieve better performance, a resonance-based band-pass lter
with 4 MHz bandwidth and 10% tuning range across the center
frequency of 1.15 GHz with the aid of adjustment screw, is used
in the receive path to lter undesired signals. The structure of
this lter is shown in Fig. 5(a). A metallic shaft is placed in a
cylindrical tube to form a coaxial transmission line. The transmission line is grounded from one side and has inductive characteristics since its length is less than
. On the other side,
there are two-adjustment screws that form a capacitance with
the internal metallic shaft. The dimensions of the external tube
and the internal metallic shaft are chosen to form a coaxial transmission line with
characteristic impedance for the highest
available quality factor. By utilizing this simple structure, an
ultra-high quality factor (
) band-pass lter
can be implemented at an extremely low cost. Fig. 5(b) shows
the measured insertion loss of the fabricated BPF.
The third block consists of a 0-360 phase shifter and
attenuators (HMC307QS16G). Both shifter and attenuator
are controlled digitally. The phase shifter with 0.1 /step is
implemented using an analog phase shifter and a digital to
analog converter (D/A). The analog phase shifter is designed

12

IEEE EMBEDDED SYSTEMS LETTERS, VOL. 8, NO. 1, MARCH 2016

TABLE I
HORN ANTENNA VERSUS MICROSTRIP PATCH ANTENNA

Fig. 6. The proposed horn antennas: (a) 3-D radiation pattern; (b) return loss;
(c) 2-D radiation pattern; and (d) efciency.

with nonlinear transmission lines using voltage-controlled

reverse-mode varactor diodes. The ne attenuator consists of
a PIN diode-based (HSMP3814) voltage-controlled attenuator
and a D/A with 0.1 dB attenuation steps. Varactors used in
analog phase shifter can produce harmonics if the attenuation is
low in the signal path. A microstrip-based low pass lter with
1.2 GHz cut-off frequency can be utilized to eliminate these
undesired harmonics.
The fourth block consists of a mixer (LRMS-5 H), a digitally-controlled 0180 phase shifter, RF amplier, and a low
pass lter. Both transmit and receive paths have same aluminum
horn antennas with 11.5 dB free space gain at 1.15 GHz center
frequency. The horn antenna has a wideband structure thus, its
return loss near the earthquake rubble is nearly constant. Also,
antennas receiving factor is independent of the distance between antenna and the rubble. Fig. 6 shows the designed horn
antennas 3-D radiation pattern, return loss, 2-D radiation pattern and efciency.
The microstrip antenna with 7 dB free space gain is utilized in
the receive path in order to compare with horn antenna. Calculations and measurements show that required microstrip antenna
must have large dimensions ( cm
cm) for better efciency. Large dimensions introduce background noise (walking
interference) that signicantly degrades operation of the detector. Algorithms like horn antenna help in eliminating this
noise [5]. Table I lists benets of horn antenna over microstrip.
Direct coupling between transmitter and receiver has been reduced by utilizing two separate antennas. Coupling can also
be seen as clutter and its effect is removed in the clutter cancelling circuit. If the coupling is high and clutter cancelling circuit is unable to perform properly, antennas can be distanced
from each other. Through experiment it was observed that distance of 50 cm between both antennas is acceptable. One of the
key features of this design is simple structure of the demodulator. Demodulator consists of a mixer which has an LO signal
of
and an RF signal of
, where
is the radian frequency of the transmitted signal, is a constant
phase and
is the radian Doppler frequency. Mixer output
after low pass ltering is proportional to
. Phase
shifter in the LO path of mixer makes the , phase difference

Fig. 7. Simulated earthquake rubble utilized to test the proposed device.

Fig. 8. Environment noise received with nobody inside the cavity: (left) time
domain; (right) spectrum.

between RF and LO, nearly zero to increase sensitivity. To detect the Doppler frequency shift due to a human body vibration
(breathing and heart pumping), one can take Fourier transform
of the sampled output mixer. The near carrier phase noise of the
transmitting signal and the low frequency icker noise of the
diodes in the mixer can mask the output signal and set the limit
for detection. Use of low phase noise transmitting signal and
low icker noise Schottky barrier diodes in the double balance
mixer increases performance of the life detection system.
III. EXPERIMENTAL RESULTS
Fig. 7 demonstrates the simulated environment where a
human is trapped beneath the rubble. Thickness of the rubble
can change up to 1.5 m. To detect the victim, one antenna should
be placed at the top of the rubble and other antenna should
be kept near the rubble. Surrounding area can be scanned by
changing the direction of antennas. In the rst stage, the system
should be calibrated in order to eliminate clutter in the reected
wave. The system was rst tested without any human subjects
with results shown in Fig. 8.
By manipulating the information with nobody inside the
cavity, the system is calibrated and ready to use for nding
human subjects. Fig. 9 shows the reected waves and essential
signs of a middle-aged male under the cavity of thicknesses
100 cm. Figures on the left shows received signal in time
domain and gures on right are the fast Fourier transform (FFT)
of the measured signals. As shown in the frequency-domain
plots of Fig. 9, the breathing signal has a 0.35 Hz fundamental

JALALIBIDGOLI et al.: A COMPACT PORTABLE MICROWAVE LIFE-DETECTION DEVICE FOR FINDING SURVIVORS

13

IV. CONCLUSION
A simple structured, ultra-sensitive Doppler effect-based
portable life detection system with 1.15 GHz operating frequency has been introduced and examined with promising
results. Use of two antennas eliminate the need for a circulator
and simplies the search process of a victim by changing
the position of the receiving antenna. Also, a simple clutter
canceling circuit and background noise algorithm is used.
The system is able to detect human subjects trapped under
earthquake rubble with thicknesses up to 1.5 m in replicated
test environment and 10 m in realistic situations. The device
was operated with a dry-cell battery to demonstrate portability.

Fig. 9. Detected breathing and heartbeat with their spectrum under 70 cm thickness with subject breathing normally (up, left); time domain (up, right); spectrum and 80 cm thickness with subject asked to hold his breath for 3 periods
each lasts 30s (down, left); time domain (down, right) Spectrum.

Fig. 10. Detected breathing signal and its spectrum with 150 cm thickness
(left); time domain (right); spectrum.

Fig. 11. Detected breathing signal and its spectrum with 10 m thickness: (left)
time domain; (right) Spectrum.

tone and heartbeat has a 1.4 Hz fundamental tone. The other

peaks are due to harmonics, mixing products and nonlinear
effects. Measurement and simulation results show that it is
much easier to detect a trapped human under the earthquake
rubble. Fig. 10 shows the similar test under a 150 cm thickness. The baseband signal-to-noise ratios (SNRs) of proposed
detector through 100 and 150 cm rubble are approximately 55
and 50 dB, respectively. These SNRs are high enough to detect
desired signals.
Fig. 11 shows the reected waves and important signs of a
middle-aged male in realistic situations under the thicknesses
around 10 m as claimed before. To the best of authors knowledge, this is the highest reported thickness that has ever reported
in any published papers.

REFERENCES
[1] H. R. Chuang, H. C. Kuo, F. L. Lin, T. H. Huang, C. S. Kuo, and Y.
W. Ou, 60 GHz millimeter-wave life detection system (MLDS) for
noncontact human vital-signal monitoring, IEEE Sensors J., vol. 12,
no. 3, pp. 602609, Mar. 2012.
[2] J. C. Lin, Microwave sensing of physiological movement and volume
change: A review, Bioelectromagnetics, vol. 13, pp. 557565, 1992.
[3] K. M. Chen, D. Misra, H. Wang, H.-R. Chuang, and E. Postow, An
X-band microwave life-detection system, IEEE Trans. Biomed. Eng.,
vol. 33, no. 7, pp. 697702, Jul. 1986.
[4] H. R. Chuang, Y. F. Chen, and K. M. Chen, Automatic clutter-canceller for microwave life-detection system, IEEE Trans. Instrum.
Measur., vol. 40, no. 4, pp. 747750, Aug. 1991.
[5] K. M. Chen, Y. H. Huang, J. Zhang, and A. Norman, Microwave
life-detection systems for searching human subjects under earthquake
rubble or behind barrier, IEEE Trans. Biomed. Eng., vol. 27, no. 1, pp.
105114, Jan. 2000.
[6] K. M. Chen, Y. Huang, A. Norman, and J. Zhang, Microwave life-detection system for detecting human subjects through barriers, in Proc.
Progress Electromagn. Symp., Hong Kong, Jan. 69, 1997.
[7] Q. Zhou, J. Liu, A. H. Madsen, O. B. Lubecke, and V. Lubecke, Detection of multiple heartbeats using doppler radar, in Proc. ICASSP,
May 2006, vol. 2, pp. 11601163.
[8] L. Chioukh, H. Boutayeb, D. Deslandes, and K. Wu, Noise and sensitivity of harmonic radar architecture for remote sensing and detection
of vital signs, IEEE Trans. Microw. Theory Tech., vol. 62, no. 9, pp.
18471856, Sep. 2014.
[9] C. Li and J. Lin, Random body movement cancellation in doppler
radar vital sign detection, IEEE Trans. Microw. Theory Tech., vol. 56,
no. 12, pp. 31433153, Dec. 2008.
[10] M. Ascione, A. Buonamo, M. DUrso, L. Angrisani, and R. S. L.
Moriello, A new measurement method based on music algorithm
for through the wall detection of life signs, IEEE Trans. Instrumen.
Measur., vol. 62, no. 1, pp. 1327, Jan. 2013.
[11] C. Li, V. M. Lubecke, O. B. Lubecke, and J. Lin, A review on
recent advances in doppler radar sensors for noncontact healthcare
monitoring, IEEE Trans. Microw. Theory Tech., vol. 61, no. 5, pp.
20462061, May 2013.
[12] W. Hu, Z. Zhao, Y. Wang, H. Zhang, and F. Lin, Noncontact accurate
measurement of cardiopulmonary activity using a compact quadrature
doppler radar sensor, IEEE Trans. Biomed. Eng., vol. 61, no. 3, pp.
725736, Mar. 2014.
[13] Y. Hong, S. G. Kim, B. H. Kim, S. J. Ha, H. J. Lee, G. H. Yun, and
J. G. Yook, Noncontact proximity vital sign sensor based on PLL for
sensitivity enhancement, IEEE Trans. Biomed. Circuits Syst., vol. 8,
no. 4, pp. 584594, Aug. 2014.
[14] G. Wang, J. M. Ferreras, C. Gu, C. Li, and R. G. Garcia, Application
of linear frequency modulated continous-wave (LFMCW) radars for
tracking of vital signs, IEEE Trans. Microw. Theory Tech., vol. 62,
no. 6, pp. 13871400, Jun. 2014.
[15] C. Li, J. Cummings, J. Lam, E. Graves, and W. Wu, Radar remote
monitoring of vital signs, IEEE Microw. Mag., vol. 10, no. 1, pp.
4756, Feb. 2009.
[16] Q. Aardal, Y. Paichard, S. Brovoll, T. Berger, T. S. Lande, and S. E.
Harman, Physical working principles of medical radar, IEEE Trans.
Biomed. Eng., vol. 60, no. 4, pp. 11421150, Apr. 2013.