Wideband measurement strategies: from RADAR to passive wireless sensors ... and how passive wireless sensors were/are used by intelligence agencies.

Texte intégral

(1)Passive sensor interrogation J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature. Wideband measurement strategies: from RADAR to passive wireless sensors ... and how passive wireless sensors were/are used by intelligence agencies.. Demonstration using RF-filter GNURadio demonstration. J.-M Friedt, G. Goavec-Mérou. Conclusion & perspectives Appendices. FEMTO-ST Time & Frequency/SENSeOR jmfriedt@femto-st.fr slides and references available at http://jmfriedt.free.fr/. January 31, 2016 1 / 46.

(2) Passive sensor interrogation. Introduction. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives. • Passive transducers acting as RADAR cooperative targets for. sensing purposes • No local energy source, impossible to detect unless remotely. powered • “High” bandwidth measurement (> 16 kHz for voice reconstruction). environment. Appendices. sensor. emitter. receiver. 2 / 46.

(3) Passive sensor interrogation. Introduction. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature. • Passive transducers acting as RADAR cooperative targets for. sensing purposes • No local energy source, impossible to detect unless remotely. powered. Demonstration using RF-filter GNURadio demonstration. • “High” bandwidth measurement (> 16 kHz for voice reconstruction). Conclusion & perspectives. environment sensor. Appendices. emitter. 1. 5 us. receiver. How to introduce a dedicated signature from a cooperative target ?. 3 / 46.

(4) Passive sensor interrogation. Introduction. J.-M Friedt Introduction Background. • Passive transducers acting as RADAR cooperative targets for. Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration. sensing purposes • No local energy source, impossible to detect unless remotely. powered • “High” bandwidth measurement (> 16 kHz for voice reconstruction). environment. Conclusion & perspectives. sensor. Appendices. emitter. 5 us. receiver. 1. How to introduce a dedicated signature from a cooperative target ?. 2. How to add a sensing capability ? 4 / 46.

(5) Passive sensor interrogation. Introduction. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter. • Passive transducers acting as RADAR cooperative targets for. sensing purposes • No local energy source, impossible to detect unless remotely. powered • “High” bandwidth measurement (> 16 kHz for voice reconstruction). GNURadio demonstration. environment sensor. Conclusion & perspectives Appendices. emitter. 5 us. receiver. 1. How to introduce a dedicated signature from a cooperative target ?. 2. How to add a sensing capability ?. 3. How to separate the emitted signal from the returned signal ? 5 / 46.

(6) Passive sensor interrogation. Historical background: RADAR cooperative targets. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature. 1. target whose backscattered signal is representative of its state (identification, measurement). 2. active targets: radar beacons (racon), IFF. 3. passive targets: buried dielectric reflectors, Lüneberg spheres this work: use of radiofrequency transducers based on surface acoustic wave propagation (RF filters). Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 4. H. Stockman, Communication by means of reflected power Proc. IRE 36 (Oct. 1948) pp.1196–1204 (picture from http://geogdata.csun.edu/~aether/pdf/volume_05a/rosol.pdf). 6 / 46.

(7) Passive sensor interrogation. Historical background: RADAR cooperative targets. J.-M Friedt Introduction Background Passive sensor strategy. 1. target whose backscattered signal is representative of its state (identification, measurement). 2. active targets: radar beacons (racon), IFF. 3. passive targets: buried dielectric reflectors, Lüneberg spheres. 4. this work: use of radiofrequency transducers based on surface acoustic wave propagation (RF filters). Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. A. Glinsky, Theremin: Ether Music And Espionage, University of Illinois Press (2005) P. Wright & P. Greengrass, Spycatcher (1987), pp.14–17. ⇒ MI5 SATYR. http://madmikesamerica.com/2010/08/the-thing-and-the-curious-life-of-leon-theremin/thing2/. 7 / 46.

(8) Passive sensor interrogation. Historical background: RADAR cooperative targets. J.-M Friedt Introduction Background Passive sensor strategy. 1. target whose backscattered signal is representative of its state (identification, measurement). GNURadio demonstration. 2. active targets: radar beacons (racon), IFF. Conclusion & perspectives. 3. passive targets: buried dielectric reflectors, Lüneberg spheres. 4. this work: use of radiofrequency transducers based on surface acoustic wave propagation (RF filters). Examples in the literature Demonstration using RF-filter. Appendices. J. Appelbaum, J. Horchert & C. Stöcker, Shopping for Spy Gear: Catalog Advertises NSA Toolbox, Der Spiegel (12/29/2013) http://leaksource.info/2013/12/30/ nsas-ant-division-catalog-of-exploits-for-nearly-every-major-software-hard. 8 / 46.

(9) Passive sensor interrogation. Historical background: RADAR cooperative targets. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature. 1. Demonstration using RF-filter. target whose backscattered signal is representative of its state (identification, measurement). “we received the radar’s signal and fed it into a variable delay line before transmitting the signal back to the radar. By smoothly varying the length of the delay line, we could simulate the. GNURadio demonstration. 2. active targets: radar beacons (racon), IFF. Conclusion & perspectives. 3. passive targets: buried dielectric reflectors, Lüneberg spheres. 4. this work: use of radiofrequency transducers based on surface acoustic wave propagation false target’s range and (RF filters). Appendices. G. Poteat, Stealth, countermeasures and ELINT, 1960–1975 (U) (2014). speed. ... we could now simulate an aircraft of any radar cross section ”. 9 / 46.

(10) Passive sensor interrogation. Historical background: RADAR cooperative targets. J.-M Friedt Introduction Background Passive sensor strategy. 1. target whose backscattered signal is representative of its state (identification, measurement). 2. active targets: radar beacons (racon), IFF. 3. passive targets: buried dielectric reflectors, Lüneberg spheres this work: use of radiofrequency transducers based on surface acoustic wave propagation (RF filters). Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 4. D. J. Thomson, D. Card, and G. E. Bridges, RF Cavity Passive Wireless Sensors With Time-Domain Gating-Based Interrogation for SHM of Civil Structures, IEEE Sensors Journal . 9 (11) (Nov. 2009), pp.1430-1438. 10 / 46.

(11) Passive sensor interrogation. Historical background: RADAR cooperative targets. J.-M Friedt Introduction Background Passive sensor strategy. 1. target whose backscattered signal is representative of its state (identification, measurement). GNURadio demonstration. 2. active targets: radar beacons (racon), IFF. Conclusion & perspectives. 3. passive targets: buried dielectric reflectors, Lüneberg spheres. 4. this work: use of radiofrequency transducers based on surface acoustic wave propagation (RF filters). Examples in the literature Demonstration using RF-filter. Appendices. C.T. Allen, S. Kun, R.G Plumb, The use of groundpenetrating radar with a cooperative target, IEEE Transactions on Geoscience and Remote Sensing, 36 (5) (Sept. 1998) pp. 1821– 1825. 11 / 46.

(12) Passive sensor interrogation. Development strategy. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. • Electromagnetic to acoustic wave conversion for shrinking sensor. size: 300 m/µs → 3000 m/s (piezoelectric substrate) • Delay line for delayed echo. (1 µs=1.5 mm acoustic path). returned power (dB). 1. • Resonators for energy. storage and slow release. reflected signals and clutter. 2. • 20 log10 (e) = 8.6 and. exponential decay of an unloading resonator is exp(−t/τ ) with τ = Q/(π · f0 ). resonator response −8.6 dB/τ delay−line response noise t. • Rising frequency for shrinking. antenna size: λ (m) = 300/f (MHz). sensor measurement. 1 L. Reindl & al., Theory and application of passive SAW radio transponders as sensors., IEEE Trans. Ultrasonics, Ferroelectrics, and Frequency Control 45 (5), 1281–1292 (1998) 2 https://www.isa.org/participate-in-a-technical-division/communicationsdivision/dielectric-resonators 12 / 46.

(13) Passive sensor interrogation. RFID v.s linear transducers. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter. RFID. linear transducers. rectifier for powering the microcontroller = diodes. linear transduction process (piezoelectric, dielectric resonator loading). near field coupling. far field interrogation. backscattered signal modulation. delayed echo. periodic bit sampling. frequency sweep or Fourier transform of re-. GNURadio demonstration Conclusion & perspectives. turned signal off resonance continuous wave powers the microcontroller. Appendices. at resonance amplitude modulated backscattered signal rectifier microprocessor. S. Preradovic & N.C. Karmakar, Multiresonator-Based Chipless RFID Barcode of the Future, Springer (2012). “Chipless RFID” ! 13 / 46.

(14) Passive sensor interrogation. Delay line design. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives. interdigitated electrodes patterned on a piezoelectric substrate and connected to the antenna generate a radiofrequency acoustic pulse 2 the acoustic pulse, confined to the substrate surface, propagates at a few km/s ( 300 m/µs), 3 electrodes patterned on the acoustic path act as mirrors 4 echos are returned to the RADAR with a delay dependent on acoustic velocity (∝ T, σ ...) Drawback: electromechanical conversion efficiency (a few % at most) 1. Appendices. τ. τ t 0 14 / 46.

(15) Passive sensor interrogation J.-M Friedt Introduction Background. How to use a transducer for measurement ? Intrinsic material property. Passive sensor strategy. • acoustic velocity with temperature, stress: c(T , σ) =. Examples in the literature. • boundary condition (gravimetric sensor):. Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. q. E ρ. sensitivity ' (attached mass)/(moving mass) ⇒ for a given layer thickness, dependent on λ & wave confinement • a dielectric resonator resonance frequency is a 34 + 3.45 GHz (a radius, L length in mm) 3 ⇒ fTE 01 ' a√ εr L dilatation (dfTE 01 '. 34 dL √ εr L2 ). and permettivity effect. Extrinsic transducer property • varying load on a 2-port device (acoustic coupling between port connected to antenna and port connected to sensor): resistor (strain gauge), capacitor (moisture), light sensitive diode 4 • disturbing the fringing electrical field 5 3 D.. Kafjez & P. Guillon, Dielectric resonators, 2nd Ed, Noble (1998), p.3. 4 http://www.google.com/patents/US8339219 5 K. Stroganov, V. Pashchenko, V. Kalinin, V. Plessky, Passive Wireless SAW MEMS Pressure sensor/Microphone, Proc 2014 EFTF 15 / 46.

(16) Passive sensor interrogation. Application examples. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 6 6 Wang & al., Wireless surface acoustic wave chemical sensor for simultaneous measurement of CO2 and humidity, J. Micro/Nanolith. MEMS MOEMS 8 (3) (2009) 16 / 46.

(17) Passive sensor interrogation. Application examples. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 6 6 D.J. Thomson & al., RF Cavity Passive Wireless Sensors With Time-Domain Gating-Based Interrogation for SHM of Civil Structures, IEEE Sensors 9 (11), 2009 17 / 46.

(18) Passive sensor interrogation. RADAR measurement. J.-M Friedt Introduction. • Challenge of RADAR measurement: detect weak backscattered. Background. Examples in the literature. signal while the strong emitted signal makes the receiver blind • Objective: differentiate the returned signal from the emitted signal (isolation). Demonstration using RF-filter. • Frequency domain: sweeping. Passive sensor strategy. GNURadio demonstration Conclusion & perspectives Appendices. the emitted frequency so the returned signal at t + τ is at a different frequency than the emitted signal at time t + τ : beat signal at df = ∆T ∆f ∆f · τ • Doppler RADAR: only velocity of target induces frequency shift, no sweep (CW) • Time domain: separate. emission and reception. f. τ. FFT. df τ t ∆T. G. L. Charvat, Small and Short Range RADAR Systems, CRC Press (2014). In all cases: spatial (temporal) resolution given by the inverse of the bandwidth (∆R = c/(2∆f )) 18 / 46.

(19) Passive sensor interrogation. RADAR measurement. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. • Challenge of RADAR measurement: detect weak backscattered. signal while the strong emitted signal makes the receiver blind • Objective: differentiate the returned signal from the emitted signal. (isolation) • Frequency domain: sweeping. the emitted frequency so the returned signal at t + τ is at a different frequency than the emitted signal at time t + τ : beat signal at df = ∆T ∆f ∆f · τ • Doppler RADAR: only velocity of target induces frequency shift, no sweep (CW) • Time domain: separate. emission and reception. τ. FFT. f. τ df 0 6= df df τ t ∆T. In all cases: spatial (temporal) resolution given by the inverse of the bandwidth (∆R = c/(2∆f )) 19 / 46.

(20) Passive sensor interrogation. RADAR measurement. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. • Challenge of RADAR measurement: detect weak backscattered. signal while the strong emitted signal makes the receiver blind • Objective: differentiate the returned signal from the emitted signal. (isolation) • Frequency domain: sweeping. the emitted frequency so the receiver returned signal at t + τ is at a WB stroboscopy different frequency than the power emitted signal at time t + τ : ∼ 1/∆f ∆T beat signal at df = ∆f · τ. τ. • Doppler RADAR: only velocity. of target induces frequency shift, no sweep (CW) • Time domain: separate. τ. t emission and reception In all cases: spatial (temporal) resolution given by the inverse of the bandwidth (∆R = c/(2∆f )) 20 / 46.

(21) Passive sensor interrogation. Software defined radio. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. • Time gating: no need to record signal before sensor response +. focus on sensor response • Flexibility in sensor response tracking: digital version of the lock-in. amplifier BUT • need to synchronize emission and reception • low A/D resolution reduces interrogation range • dedicated hardware (DDS + switch + power detector) • low bandwidth of the DVB-T receiver (2.7 MHz ⇒ ∆R ∼ 56 m) Low cost targets 7 to play with: • TV dish receiver dielectric resonator • SAW 8 resonator – narrowband device • SAW filters – must be selected to match emitter spectrum 7 M. Ossman, The NSA Playset: RF Retroreflectors, DEF CON 22, available at https://www.youtube.com/watch?v=mAai6dRAtFo, and http://www.nsaplayset.org/ 8 Surface Acoustic Wave 21 / 46.

(22) Passive sensor interrogation. Experimental demonstration. J.-M Friedt Introduction Background. 1. Passive sensor strategy Examples in the literature Demonstration using RF-filter. A SAW filter delays the incoming electromagnetic signal by 2.7 µs and multiples ⇒ pulse repetition rate 100 kHz. 1/14/2016 6:56:14 PM 1311.6010K42-103899-La. Trc1. S11 dB Mag 1 dB/ Ref -4 dB. Cal Oﬀ. 1. 1 0 -1 -2 -3. -4-4dB -5 -6 -7 -8 -9. GNURadio demonstration Conclusion & perspectives Appendices. 2. pulse duration: 2 µs to separate echos and improve SNR of 2.7 MS/s DVB-T receiver signal (5-6 samples/pulse). Ch1 Center 140 MHz Trc3. Pwr -10 dBm Bw 10 kHz. S11 dB Mag 10 dB/ Ref -50 dB. 0. Span 10 MHz. Cal Oﬀ. 2 M1 2.706 µs -16.7198 dB. M1. -10. 2.7 us. -20. 2x2.7 us =5.4 us. -30 -40. 3x2.7us =8.1 us. -50dB -50 -60 -70 -80 -90 -100. Ch1 Center 140 MHz Trc3 Start. 100 ns. Pwr -10 dBm Bw 10 kHz Time Domain. Span 10 MHz Stop 10 µs. TDK/Epcos B3607 140 MHz filter (5 euros on Ebay) network analyzer characterization: top S11 (freq. domain), bottom S11 (time domain). 22 / 46.

(23) Passive sensor interrogation. Experimental demonstration. J.-M Friedt Introduction Background. 1. Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. A SAW filter delays the incoming electromagnetic signal by 2.7 µs and multiples ⇒ pulse repetition rate 100 kHz. 1/13/2016 3:30:44 PM 1311.6010K42-103899-La. Trc1. S11 dB Mag 1 dB/ Ref -3 dB. Cal int. 1. 2. M1 140.1796 MHz -1.9056 dB M2 146.7066 MHz -0.5729 dB. 1. M2. 0 -1. M1. -2. -3-3dB -4 -5 -6. 2. pulse duration: 2 µs to separate echos and improve SNR of 2.7 MS/s DVB-T receiver signal (5-6 samples/pulse). -7 -8. Ch1 Center 140 MHz. Pwr -10 dBm Bw 10 kHz. Trc3. Cal int Gat. S11 dB Mag 10 dB/ Ref -40 dB. M1 140.1796 MHz -16.3012 dB M2 146.7066 MHz -44.1265 dB. 0 -10. Span 20 MHz 2. 10. M1. -20 -30. M2. -40dB -40 -50 -60 -70 -80 -90. Ch1 Center 140 MHz. Pwr -10 dBm Bw 10 kHz. Span 20 MHz. TDK/Epcos B3607 140 MHz filter (5 euros on Ebay) network analyzer characterization: top S11 , bottom time-gated between 2 and 3 µs. 23 / 46.

(24) Passive sensor interrogation. Experimental demonstration. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature. 1. A SAW filter delays the incoming electromagnetic signal by 2.7 µs and multiples ⇒ pulse repetition rate 100 kHz. 2. pulse duration: 2 µs to separate echos and improve SNR of 2.7 MS/s DVB-T receiver signal (5-6 samples/pulse). Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 3. signal reflected from the filter (impulse response). open v.s 50 Ω load. 24 / 46.

(25) Passive sensor interrogation. Experimental demonstration. J.-M Friedt Introduction Background. 1. Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration. 2. Conclusion & perspectives Appendices. 3. A SAW filter delays the incoming electromagnetic signal by 2.7 µs and multiples ⇒ pulse repetition rate 100 kHz pulse duration: 2 µs to separate echos and improve SNR of 2.7 MS/s DVB-T receiver signal (5-6 samples/pulse). 2.7 µs PRR= 100 kHz C switch 1 2 140 MHz. Z. DVB−T P. Emit. Emit Receive. Emit Receive t. variable capacitor (10-100 pF) MAX2606 VCO as source, Hittite selected to induce a 50 Ω load switch, R820T2 receiver, PWM outat operating pulsation ω, and put of Atmega32U4 for switching inductor (56 nH) to cancel the imaginary component. 25 / 46.

(26) Passive sensor interrogation. Experimental demonstration. J.-M Friedt Introduction Background Passive sensor strategy. 1. Examples in the literature Demonstration using RF-filter GNURadio demonstration. 2. Conclusion & perspectives Appendices. 3. A SAW filter delays the incoming electromagnetic signal by 2.7 µs and multiples ⇒ pulse repetition rate 100 kHz pulse duration: 2 µs to separate echos and improve SNR of 2.7 MS/s DVB-T receiver signal (5-6 samples/pulse) the digital signal to be observed controls a FET (BF996S) which connects the filter to a 50 Ω load. C~1/(50. ω). 2. L~1/(C. ω ) 2.7 µs. PRR= 100 kHz 140 MHz. DVB−T P. Emit. Emit C1. Emit C2. C3 t. 26 / 46.

(27) Passive sensor interrogation. Experimental demonstration. J.-M Friedt Introduction Background. 11 00 00 11 00 11 00 11 00 11 00 11 00 11 00 11. Passive sensor strategy Examples in the literature. 1. Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 2. A SAW filter delays the incoming electromagnetic signal by 2.7 µs and multiples ⇒ pulse repetition rate 100 kHz pulse duration: 2 µs to separate echos and improve SNR of 2.7 MS/s DVB-T receiver signal (5-6 samples/pulse). 0000000 1111111 0000000 1111111 1111111111111111 0000000000000000 0000000RS232 00000000000000001111111 1111111111111111 digital control. 2.7 µs PRR= 100 kHz 140 MHz. 50. DVB−T P. Emit. Emit ’1’. Emit ’0’. ’1’ t. 27 / 46.

(28) Passive sensor interrogation. Experimental demonstration. J.-M Friedt. 0.7. Introduction. 0.6. Passive sensor strategy. 0.5. Examples in the literature. 1. Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 2. A SAW filter delays the incoming electromagnetic signal by 2.7 µs and multiples ⇒ pulse repetition rate 100 kHz pulse duration: 2 µs to separate echos and improve SNR of 2.7 MS/s DVB-T receiver signal (5-6 samples/pulse). |I+jQ| (a.u.). Background. first echo (for measurement). 0.4. 0.3. 0.2. 0.1. 0. 0. 100 second echo. 200. 300. time (2.7 MS/s). 400. emission500. 28 / 46.

(29) Passive sensor interrogation. Demonstration using an acoustic filter. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration. 1. only keep the returned signal above a given threshold. 2. 9600 baud ⇒ 1/9600 s/symbol=104 µs/symbol (start, 8 bits, stop). 3. 10 pulses/bit and 5 samples/pulse ⇒ 50 I/Q samples/RS232 bit. 4. sliding average on 15 samples to keep sharp transitions from 1 to 0. Conclusion & perspectives 0.8. 0.8. 0.6. 0.6. 0.4. 0.4. 0.2. 0.2 signal (a.u.). signal (a.u.). Appendices. 0. 0. -0.2. -0.2. -0.4. -0.4. -0.6. -0.6. -0.8 0. 5000. 10000. 15000 20000 time (us). 25000. 30000. 35000. -0.8 3000. 4000. 5000. 6000. 7000. 8000. 9000. 10000. time (us). 29 / 46.

(30) Passive sensor interrogation. Demonstration using an acoustic filter. J.-M Friedt Introduction Background. Demonstration using RF-filter. 2 3 4. only keep the returned signal above a given threshold 9600 baud ⇒ 1/9600 s/symbol=104 µs/symbol (start, 8 bits, stop) 10 pulses/bit and 5 samples/pulse ⇒ 50 I/Q samples/RS232 bit sliding average on 15 samples to keep sharp transitions from 1 to 0. GNURadio demonstration. 1. MSB stop. Examples in the literature. 1. Start LSB. Passive sensor strategy. Conclusion & perspectives Appendices. signal (a.u.). 0.5. 0. −0.5 S00010010s S10100110s S00110110s S00110110s S11110110s S01010000s. H −1 8000. 9000. e. l. l. o. \n. 10000 11000 12000 13000 14000 15000 16000 17000 time (us) 30 / 46.

(31) Passive sensor interrogation J.-M Friedt Introduction Background. GNURadio real time decoding Continuously recover the signal including emission and reception times (2.7 MS/s). Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 31 / 46.

(32) Passive sensor interrogation J.-M Friedt Introduction Background. GNURadio real time decoding Threshold defines the returned echo while low level is observed during emission. Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 32 / 46.

(33) Passive sensor interrogation J.-M Friedt Introduction Background. GNURadio real time decoding Issue of variable rate: echoes don’t always last the same duration ⇒ peak holder (green), with encoding signal visible. Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 33 / 46.

(34) Passive sensor interrogation. GNURadio real time decoding. J.-M Friedt Introduction Background Passive sensor strategy. Threshold on the detected signal and low pass filtering for decimation (2700 →100 kS/s) ⇒ at 9600 bauds (104 µs/bit), 10 samples/bit ⇒ decoding independent on PRR9. Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 9 Pulse. Repetition Rate 34 / 46.

(35) Passive sensor interrogation J.-M Friedt. GNURadio real time decoding. Introduction Background. Real time decoding of the recorded message (start – 8 bits – stop). Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 35 / 46.

(36) Passive sensor interrogation. Membrane design. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 9. Membrane geometry : 2 Radius a = 1 cm, gap h = 50 · 10−6 m ⇒ C = ε0 εr π·a h = 56 pF Tension load N0 = 2000 N/m along circumference 10 h Pressure p0 =1 Pa (traffic on a busy roadway) ⇒ displacement W0 at center of membrane by considering force equilibrium: tension=pressure (assuming negligible membrane stiffness and inertia) 2 0 ·a p0 × πa2 = 4πN0 × W0 ⇔ W0 = p4N = 125 nm and 0 δC C. =. W0 h. = 0.25% or δC = 1 pF. 2a. p0 W0. (×10 if N0 divided by 10 since sound quality does not matter) 9 Tradeoff between acoustic and RF characteristics: reentrant cavity design proposed in Report on Research on EASYCHAIR (1955) at http://www.cryptomuseum.com/covert/bugs/ec/files/19550714_cia.pdf 10 http://etd.dtu.dk/thesis/249608/Arnaud_DESSEIN.pdf. 36 / 46.

(37) Passive sensor interrogation. Conclusion. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature. • RADAR passive cooperative targets for sensing & tagging, • need to separate emitted pulse from returned signal: frequency •. Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives. • •. Appendices. • •. sweep (FMCW) or time gating (pulsed radar) Pulse Repetition Rate defined by echo delay, pulse length defined by recorder bandwidth ⇒ make decoding independent on PRR need to match emitter, transducer and receiver bandwidths: the narrowest part defines the whole system bandwidth demonstrated with acoustic filter and PAL delay line (sensing principle ?) demonstrated response modulation by varying load under digital control (RS232) and recovering the signal from returned RF signal Demonstration using the Redpitaya for >500 kS/s refresh rate 11 , now with gnuradio support in buildroot12 .. 11 G. Goavec-Mérou & al., Fast contactless vibrating structure characterization using real time FPGA-based digital signal processing: demonstrations with a passive wireless acoustic delay line probe and vision, Rev. Sci. Instrum. 85 (1), Jan. 2014, pp.015109 12 https://github.com/trabucayre/redpitaya 37 / 46.

(38) Passive sensor interrogation. Perspectives. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 5.8 The RF circuit method. 13. “ Before low-noise field effect transistors were available, semiconductor technology was applied to condenser microphones in the form of the so-called radio-frequency circuit method, which requires only conventional transistors. With an RF circuit, the microphone capsule operates as an “active transducer” (see Section 2): It controls the frequency or phase of an RF oscillator or represents an impedance in an RF circuit that varies in cadence with the audio frequency.”. Perspectives: • identify a frequency source and filter for operating at higher. frequency, yet keeping the long delay, • demonstrate a wireless link (50 cm monopole) instead of simulated. propagation losses with attenuator (FSPL'70 dB @ 10 m 2-way), • coherent approach for phase exploitation (instead of magnitude). requires shared clock between emitter and receiver. Acknowledgements: V. Plessky, S. Ballandras, G. Martin, L. Reindl & T. Ostertag 13 G. Boré & S. Peus, Microphones – Methods of Operation and Type Examples (4th Ed.), (1999), p.43, initial release 1973, available at http://www.neumann.com/downloadmanager/d.php?download=docu0002.PDF 38 / 46.

(39) Passive sensor interrogation. PAL delay line. J.-M Friedt Introduction Background Passive sensor strategy. 1. 61 µs delay (too large). 2. low operating frequency, 3.2 MHz fundamental frequency (too low). 3. fringes due to delayed signal interference (c/(2d)). Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives. 0.5 0. Appendices. −0.5. −1.4 −1.6 −1.8. −2. −2 |S11| (dB). |S11| (dB). −1 −1.5. −2.5. mode spacing 7.6 kHz= 1/(2x61) us. −3 −3.5 −4. 0. 5. 10. −2.2 −2.4 −2.6 −2.8 −3 −3.2 −3.4 −3.6 3.1. 7.6 kHz 3.12. 3.14 3.16 frequency (MHz). 15 20 frequency (MHz). 3.18. 25. 3.2. 30. 39 / 46.

(40) Passive sensor interrogation. PAL delay line. J.-M Friedt Introduction Background Passive sensor strategy. 1. 61 µs delay (too large). 2. low operating frequency, 3.2 MHz fundamental frequency (too low). 3. harmonic: here 12.2 MHz. Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. 40 / 46.

(41) Passive sensor interrogation. Link budget. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. For a given transmitted power, propagation range, antenna gains and receiver sensitivity, range is given by the target cross-section σ. G 2 λ2 σ exp(−2αR) Pr = (α for non-air media, eg soil) Pt (4π)3 R 4 • notice the 4-th power of distance dependance • delay lines: σ → (G 0 )2 λ2 /(4π · IL) with IL = K 2 × A × r with A att./λ (dB/µs∝ f 2 , 0.14 dB/µs @ 300 MHz), r mirror reflection coef., K 2 = 5.4% for LNO-128, 4.8% for LTO-36. Classical IL: 30 dB, best case is ' 20 dB.. • resonators: returned power=1 − |S11 | and min(S11 ) < −10 dB. ⇒ IL ' −0.5.. − 2 dB but 8.7 dB/τ loss due to receiver delay (8.7=10· log10 (e)). K. A. K. A. r. 41 / 46.

(42) Passive sensor interrogation. Dielectric resonator wireless interrogation. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration. • Backscattered signal through farfield antenna impedance coupling. 14. • Returned signal requires fine tuning the VCO sweep rate • Here demonstrated on 9.85 GHz resonator fitted in a resonant cavity • Q ' 104 mandatory for long range measurement | S11| ( dB). Conclusion & perspectives. FSPL= +16 dBi +6 dBm −105 dB +16 dBi Siever Lab Siever Lab PM7320X PM7320X −67 dBm. Appendices. +6 dBm. 9. 81 9. 82 9. 83 9. 84 9. 85 9. 86 9. 87 9. 88 9. 89 f r equency ( GHz). 9. 9. 9. 81 9. 82 9. 83 9. 84 9. 85 9. 86 9. 87 9. 88 9. 89 f r equency ( GHz). 9. 9. 2. 2. +9 dBm Hittite HMC512. ar g( S11) ( r ad). VCO. 0 -1 -2 -3 -4 -5 -6 -7 -8 9. 8. oscilloscope 16 averages. MiniCircuits ZFL−1000+. 2 1. 8 1. 6 1. 4 1. 2 9. 8. 14 J.-M. Friedt & al., Probing a dielectric resonator acting as passive sensor through a wireless microwave link, Rev. Sci. Instrum. 85 (9), pp. 024409RSI (2014) 42 / 46.

(43) Passive sensor interrogation. Dielectric resonator wireless interrogation. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter. • • • •. Backscattered signal through farfield antenna impedance coupling 14 Returned signal requires fine tuning the VCO sweep rate Here demonstrated on 9.85 GHz resonator fitted in a resonant cavity Q ' 104 mandatory for long range measurement 0 −0.002 −0.004 −0.006 0 1400 1200 1000 800 600 400 200 0. 50. 100. 150 200 250 300 time (point index). 350. 400 2 0 −2 −4 −6. 30. 40. 50 60 70 80 temperature (deg C). 90. −8 100. temperature error (K). Appendices. 0.002. relative frequency (ppm). Conclusion & perspectives. voltage (V). 0.004. GNURadio demonstration. Using the derivate of the returned signal to convert the minimum to a zero-crossing condition. 14 J.-M. Friedt & al., Probing a dielectric resonator acting as passive sensor through a wireless microwave link, Rev. Sci. Instrum. 85 (9), pp. 024409RSI (2014) 43 / 46.

(44) Passive sensor interrogation. Membrane design++. J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. Membrane geometry: radius a, thickness h = 250 · 10−6 , tension load N0 along circumference, Material (Mylar): Young modulus E ' 4 − 5 GPa Poisson ratio ν = 0.3 Pressure p0 ⇒ displacement W0 at center of membrane D =q E · h3 /(12(1 − ν 2 )) k=. N0 ·a2 D p0 ·a4 E ·h4 15. P= (from. Chart of W0 /P as a function of k. ). 15 M. Sheplak, J. Dugundji, Large Deflections of Clamped Circular Plates Under Initial Tension and Transitions to Membrane Behavior, J. Appl. Mech. (1998) 65, 107– 44 / 46.

(45) Passive sensor interrogation J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration. GNU/Octave to GNURadio RS232 decoding r=r e a d f l o a t b i n a r y ( ’ H e l l o d e s k t o p _ 2 p 7 . bin ’ ) ; k=f i n d ( r >0.4) ; y=c o n v ( r ( k ) , o n e s ( 1 5 , 1 ) / 1 5 ) ; y=y ( 7 : end −7) ; % s l i d i n g avg plot (k (1:15000) /2.7 , y (1:15000) ) x l a b e l ( ’ time ( us ) ’ ) ; y l a b e l ( ’ signal ( a . u .) ’ ) f o r k =1:65 l i n e ( [ k ∗1 e6 /9600+3045 k ∗1 e6 / 9 6 0 0 + 3 0 4 5 ] , [ 0 . 4 0 . 5 ] ) % c l k end. Conclusion & perspectives Appendices. s =0.47 % threshold u =1; f o r k =1:4500 % l e n g t h ( k ) i f ( y ( u )>=s ) d e b u t =1; u=u+75 % 1 . 5 baud=50+25 I /Q s a m p l e s c =0; % new c h a r s t a r t b i t ( h i −>l o t r a n s i t i o n ) d e t e c t e d f o r l =0:7 i f ( y ( u )<s ) p r i n t f ( ’1 ’ ) ; c=c+2ˆ l ; e l s e p r i n t f ( ’0 ’ ) ; end u=u +50; % hardcoded baudrate : 5 samples / echo . . . end % . . . & 10 e c h o / b i t d e b u t =0; p r i n t f ( ’ %c\n’ , char ( c ) ) end u=u +1; end. 45 / 46.

(46) Passive sensor interrogation J.-M Friedt Introduction Background Passive sensor strategy Examples in the literature Demonstration using RF-filter GNURadio demonstration Conclusion & perspectives Appendices. namespace g r { namespace t h e r e m i n { [...] s e r i a l i m p l : : s e r i a l i m p l ( i n t s t e p ) : s y n c b l o c k ( " serial " , g r : : i o s i g n a t u r e : : make ( 1 , 1 , s i z e o f ( f l o a t ) ) , g r : : i o s i g n a t u r e : : make ( 0 , 0 , 0 ) ) , n e x t p o s ( 0 ) , c u r r c h a r ( ’ \0 ’ ) , s t e p ( s t e p ) , n b b i t s (0) , in frame (0) , test pos ( f a l s e ) , t o t a l (0) {} [...] v o i d s e r i a l i m p l : : f o r e c a s t ( i n t n o u t p u t i t e m s , g r v e c t o r i n t &n i n p u t i t e m s r e q u i r e d ) {} #d e f i n e HIGH STATE 1 #d e f i n e LOW STATE 0 i n t s e r i a l i m p l : : work ( i n t n o u t p u t i t e m s , g r v e c t o r c o n s t v o i d s t a r &i n p u t i t e m s , g r v e c t o r v o i d s t a r &o u t p u t i t e m s ) {c o n s t f l o a t ∗ i n = ( c o n s t f l o a t ∗) i n p u t i t e m s [ 0 ] ; v o l a t i l e long i = next pos ; i n t s t a r t s t e p = s t e p + ( s t e p >>1); unsigned char b i t ; while ( i < noutput items ) { b i t = ( unsigned char ) in [ i ] ˆ 1; i f ( i n f r a m e == 0 ) { /∗ n o t i n a f r a m e : s e a r c h f o r s t a r t b i t f a l l i n g e d g e ∗/ i f ( b i t == LOW STATE) { in frame = 1; /∗ 1/2 p e r i o d f o r s t a r t b i t + 1 p e r i o d = m i d d l e o f f i r s t b i t ∗/ i += s t a r t s t e p ; t o t a l += s t a r t s t e p ; } e l s e { i ++; t o t a l ++;} } else { i f ( n b b i t s >= 8 ) { /∗ s t o p b i t ∗/ i f ( ( c u r r c h a r >31)&&( c u r r c h a r ) ) p r i n t f ( "%c" , ( u n s i g n ed char ) c u r r c h a r ) ; e l s e p r i n t f ( " (% x ) \ n " , ( u n s i g n e d c h a r ) c u r r c h a r ) ; c u r r c h a r = ’ \0 ’ ; nb bits = 0; next pos = 0; in frame = 0; } e l s e { c u r r c h a r |= ( b i t<< n b b i t s ) ; n b b i t s ++; i += s t e p ; t o t a l+= s t e p ;} // add b i t } } i f ( i n f r a m e == 1 ) n e x t p o s = i − n o u t p u t i t e m s ; /∗ i f t r a m e i s on m u l t i p l e b u f f e r ∗/ consume each ( n o u t p u t i t e m s ) ; return noutput items ; } } /∗ namespace t h e r e m i n ∗/ } /∗ namespace g r ∗/ 46 / 46.

(47)