Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

Surface Acoustic Wave Resonators as PassiveBuried Sensors

J.-M Friedt1, T. Retornaz1, G. Martin2, T. Laroche2,J.-P. Simonnet3, E. Carry2, S. Ballandras2

1 Senseor, France2 Time & Frequency Department, FEMTO-ST, Besancon, France

3 Laboratoire de ChronoEnvironnement, Franche Comte University,Besancon, France

jmfriedt@femto-st.fr

April 21, 2009

1 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

IntoductionSurface acoustic wave (SAW) devices meet ground penetrating RADAR(GPR):

• SAW devices provide transducers able to store and release energyprovided as an electromagnetic wave ⇒ wireless

• Two major classes of transducers: narrowband resonators(frequency domain) and wideband delay lines (time domain)

• The velocity of the acoustic wave might be a function of a physicalquantity (temperature, stress ...) depending on design ⇒ signaturecharacteristic of the measured quantity

• GPR is a classical time-domain technique for probing dielectricinterfaces in ground

• bistatic antenna configuration: a short (single) RF pulse is emittedand echos are monitored ⇒ scan to gather a map of the interfaces

The interrogation method of GPR is similar to that used to probe delaylines ⇒ use SAW device as GPR cooperative target

2 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

Buried SAW resonators• Narrowband SAW resonator designed as temperature sensors are

buried at various depths ranging from 30 to 80 cm.• Frequency sweep using a narrowband RADAR to identify the

resonance frequency and hence the temperature

interrogationunit

waveguide

(conducting

wire)

T sensor

clay

30 cm

• The emitted power is 10 dBm, receiver detection limit -70 dBm ⇒in this experiment, no direct signal was gathered when burying theSAW

• An electromagnetic waveguide is provided through a metallic wirethrown in the hole (no electrical connection on either side of thewire)

3 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

Buried SAW resonators (2)• These SAW devices were not individually calibrated ⇒ relative

temperature observation• Measurement during ' 1.25 year: SAW devices are robust to the

environment• No loss in RF signal quality over time• The undergroud signal closely matches a running average over

several weeks of surface temperatures.

0 50 100 150 200 250 300 350 400 450−15

−10

−5

0

5

10

15

20

25

30

∝ T

(o C)

temps (jours depuis 01/01/2008)

σ(30 cm)relative <30 cm>σ(60 cm)relative <60 cm>σ(80 cm)relative <80 cm><temperature Besancon>

7

Air temperatures from http://www.meteo-franche-comte.fr/ 4 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

GPR• Measurement in agreement with the expected temperature trend but

poor interrogation range because we wished to respect ISM bandregulations (50% duty cycle ⇒ mean power = 0.5 peak power).

• GPR is classically used in geophysics to probe dielectric interfacesseveral meters deep

⇒ switch from resonator to delay line as GPR cooperative target 5 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

SAW delay line• Point-like reflectors appear as hyperbola• The interface echo from the buried antenna will be followed by

reflections within the delay line ⇒ the hyperbola associated withdielectic interface reflections will be followed by acoustic sensorechos.

• Challenge: GPR hardware must be adapted because a single RFpulse is insufficient to load energy in SAW delay line

6 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

SAW delay line interrogatorprototype (2)

We experimentally demonstrate that the best response of the delay lineis achieved when the number of periods of the signal is equal to thenumber of interdigitated transducers

• Tradeoff between loading a delay line (multiple periods) and spatialresolution (single pulse) ⇒ development of a dedicated solution,using a 860 MHz SAW delay line (sensor from the KongsbergSentry temperature monitoring system)

• No ISM band regulations (wideband, here 30 MHz at 860 MHz),peak power � mean power

7 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

SAW delay line interrogationprototype

Experimental results using the Universal Software Radio Peripheral(USRP, http://www.ettus.com/) and GNURadio opensource software:prototyping tools for developing RF interrogation units (opensource,modular)

USRP prototyping toolset

resonatorSAW

antennaconfiguration

monostatic

communicationUSB

transceiver433 MHz

transceiver860 MHz

FPGA

6V DC

8 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

SAW delay line interrogationprototype (3)

0

200

400

600

800

1000

1200

1400

10 20 30 40 50 60 70 80

|IQ|

time (15 ns/point)

-2

-1

0

1

2

3

4

10 20 30 40 50 60 70 80

φ IQ

time (15 ns/point)

1234

800

1000

1200

1400

1600

1800

0 200 400 600 800 1000 1200

time

(1.5

ns/

poin

t)

time (repetition rate)

1st peak2nd peak3rd peak4th peak<4th>10

• Magnitude and phase of the reflections of the delay line: the phaseis defined but does not vary, probably because the received signal istoo weak (insufficient receiver amplification)

• Signal processing: polynomial fit of the reflection magnitude andidentification of the position of the maximum referenced to theemission pulse.

9 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

SAW delay line temperaturemeasurement

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000

1000

1500

2000

2500

time (15 s/point)

dela

y (a

.u.)

reflection1reflection2reflection3reflection4temperature(degC)x10+800

−20 0 20 40 60 80 100 120 140 160

1000

1500

2000

2500

temperature (degC)

dela

y (a

.u.)

• Evolution of the duration between excitation pulse and reflections(4 reflections) as a function of temperature.

• These data are extracted from the magnitude only (' ±3oC), sincethe receiver gain was insufficient to define the phase.

10 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

Interrogation range estimate

How to estimate the interrogation range of SAW devices considering thedetection range of GPR used on ice-rock interface ?

Two approaches: planar wave approach (Fresnel equation, reflectioncoefficient from permittivity change) or FDTD modelling.

1 εice ' 3.1, εrock ' 5 ⇒ reflected power R =(√

εice−√

εrock√εice+

√εrock

)2

, here

-19 dB

2 Free Space Propagation Loss=20 log10 d + 20 log10 f + 32.44 (f inMHz and d in km): at 100 MHz, FSPL(2×100 m)'58 dB andFSPL(2×15 m) '42 dB.

3 2D-FDTD simulation considering a hole (air, ε = 1⇒ R ' -8 dB)in rock.

Since typical SAW insertion losses are in the -35 dB range, theinterrogation range will be reduced from 100 m to 15 m due tolower reflection coefficient of SAW delay line.

11 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

Sensor signature/2D-FDTDsimulations

ε1

ε2

Tx Rx

emission

SAW antenna

interface1

SAW signature

(a)(b)(c)(d)(e)

−100

−80

−60

−40

−20

0

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

Ele

ctric

fiel

d m

odul

us (

dB)

Time(µs)

(a) infinite radius, 5 m deep, source in ground(b) infinite radius, 5 m deep, source in air(c) 30 cm radius, 5 m deep, source in ground(d) 30 cm radius, 5 m deep, source in air(e) 30 cm radius, 10 m deep, source in ground

30 cm & 1 m diameter, 5 m deep in soil (εr = 25)

Expected signal measurement technique: the sensor appears as multiple hyperbola,

the summit of the 1st one being the sensor position and the delay to the next ones

being related to the physical quantity.12 / 13

Surface AcousticWave Resonatorsas Passive Buried

Sensors

Friedt & al.

Introductions

Buried resonators

GroundPenetratingRadar

Delay lineinterrogation

Measurementrange estimate

Conclusion

Conclusion and perspectives1 We have demonstrated that SAW devices provide rugged sensors

compatible with underground measurements.

2 We have illustrated the detection range of GPR on ice/rockinterface and verified that distances up to 100 m can be reached

3 We have demonstrated the use of a prototyping tool – USRP – fortesting RADAR interrogation stategies of delay lines

4 We have estimated the interrogation range of delay lines using GPR.

We still have to

1 demonstrate the wireless signal detection of buried 860 MHz delaylines using USRP

2 develop a SAW delay line working efficiently at 100 MHz, afrequency providing good compromise between antenna size andinterrogation range

3 finalize 3D-FDTD simulation to assess the interrogation range

Acknowledgement: GPR measurements were performed as part of the

Hydro-Sensor-FLOWS ANR Projet, under the supervision of M. Griselin and C.

Marlin.13 / 13