Acta Optica Sinica, Volume. 43, Issue 14, 1428001(2023)

Influencing Factors of IQ Demodulation Method in Distributed Acoustic Sensors

Lijuan Zhao1,2,3, Xuzhe Zhang1, Zhiniu Xu1、*, and Yonghui Chen1
Author Affiliations
  • 1School of Electrical and Electronic Engineering, North China Electric Power University, Baoding 071003, Hebei, China
  • 2Hebei Key Laboratory of Power Internet of Things Technology, North China Electric Power University, Baoding 071003, Hebei, China
  • 3Baoding Key Laboratory of Optical Fiber Sensing and Optical Communication Technology, North China Electric Power University, Baoding 071003, Hebei, China
  • show less
    Figures & Tables(28)
    Schematic of coherent detection φ-OTDR system
    Schematic of IQ demodulation
    Optical fiber RBS model
    Schematic of optical fiber RBS model under vibration
    Measured BPD noise signal distribution
    Simulation and measured signal spectrograms of optical fiber without vibration. (a) Simulation result; (b) measurement result
    Actual system demodulation results. (a) Amplitude variation curve; (b) 3D image of demodulated phase
    Comparison of demodulated waveform with applied vibration waveform
    Demodulated results of fiber under multiple vibration conditions
    Impact of sampling rate on demodulation effect (x-axis is logarithmic). (a) SNR of vibration detection at different sampling rates; (b) distortion at different sampling rates
    Comparison of demodulated and actual vibrations at different sampling rates and noise levels. (a) Sampling rate is 100 MHz,σ=0.080; (b) sampling rate is 1 GHz, σ=0.080; (c) sampling rate is 100 MHz, σ=0.010; (d) sampling rate is 1 GHz, σ=0.010
    Comparison of demodulated and actual vibrations at different ADC quantization bits and noise levels. (a) Number of ADC quantization bits is 8, σ=0.040; (b) number of ADC quantization bits is 16, σ=0.040; (c) number of ADC quantization bits is 8, σ=0.0050; (d) number of ADC quantization bits is 16, σ=0.0050
    Impact of pulse width on demodulation accuracy (x-axis is logarithmic). (a) SNR of vibration detection at different pulse widths; (b) distortion at different pulse widths
    Fitting results of the relationship between SNR of vibration detection, distortion, and pulse width under different noise levels. (a) Fitting results of SNR of vibration detection (x-axis is linear); (b) fitting results of distortion (x-axis is logarithmic)
    Comparison of demodulated and applied vibrations at different pulse widths and noise levels. (a) Pulse width is 20 ns, σ=0.040; (b) pulse width is 500 ns, σ=0.040; (c) pulse width is 20 ns, σ=0; (d) pulse width is 500 ns, σ=0
    Impact of magnification of EDFA on demodulation accuracy (x-axis is logarithmic). (a) SNR of vibration detection at different magnifications of EDFA; (b) distortion at different magnifications of EDFA
    Fitting results of the relationship between SNR of vibration detection, distortion, and magnification of EDFA under different noise levels (x-axis is logarithmic). (a) Fitting results of SNR of vibration detection; (b) fitting results of distortion
    Comparison of demodulated and actual vibrations at different magnifications of EDFA and noise levels. (a) Magnification of EDFA is 20, σ=0.080; (b) magnification of EDFA is 1000, σ=0.080; (c) magnification of EDFA is 20, σ=0.005; (d) magnification of EDFA is 1000, σ=0.005
    Impact of vibration position on demodulation accuracy (x-axis is linear). (a) SNR of vibration detection at different vibration positions; (b) distortion at different vibration positions
    Fitting results of the relationship between SNR of vibration detection, distortion, and vibration position under different noise (x-axis is linear). (a) Fitting results of SNR of vibration detection; (b) fitting results of distortion
    Comparison of demodulated and applied vibrations at different vibration positions and noise levels. (a) Vibration position is at 5.5 km, σ=0.040; (b) vibration position is at 0.5 km, σ=0.040; (c) vibration position is at 5.5 km, σ=0.005; (d) vibration position is at 0.5 km, σ=0.005
    • Table 1. Impact of ADC quantization bit on demodulation accuracy

      View table

      Table 1. Impact of ADC quantization bit on demodulation accuracy

      σNumber of ADC bitsSNR of vibration detection/dBD/10-4
      0835.12241.00
      1035.1214
      1235.1221
      1635.1221
      0.005834.17262.00
      1034.1712
      1234.1706
      1634.1703
      0.010832.60963.00
      1032.6107
      1232.6093
      1632.6088
      0.020829.76116.00
      1029.7614
      1229.7620
      1629.7628
      0.040824.841315.00
      1024.8422
      1224.8405
      1624.8438
      0.080819.529341.00
      1019.5297
      1219.5308
      1619.5316
    • Table 2. Fitting results of the relationship between SNR of vibration detection and pulse width under different noise levels

      View table

      Table 2. Fitting results of the relationship between SNR of vibration detection and pulse width under different noise levels

      σFitting curveR2SSE /dB2RMSE
      0Double exponential0.992.760.67
      0.005Double exponential0.993.530.77
      0.010Double exponential0.992.420.63
      0.020Double exponential0.992.070.59
      0.040Double exponential0.993.370.75
      0.080Double exponential0.992.130.65
    • Table 3. Fitting results of the relationship between distortion and pulse width under different noise levels

      View table

      Table 3. Fitting results of the relationship between distortion and pulse width under different noise levels

      σFitting curveR2SSERMSE
      0Linear1.0000
      0.005Double exponential0.978.35×10-93.73×10-5
      0.010Double exponential0.998.75×10-93.82×10-5
      0.020Double exponential0.995.44×10-89.53×10-5
      0.040Double exponential0.992.17×10-71.90×10-4
      0.080Double exponential0.993.51×10-72.65×10-4
    • Table 4. Fitting results of the relationship between SNR of vibration detection and magnification of EDFA under different noise levels

      View table

      Table 4. Fitting results of the relationship between SNR of vibration detection and magnification of EDFA under different noise levels

      σFitting curveR2SSE /dB2RMSE
      0Linear1.0000.67
      0.005Power0.990.00340.041
      0.010Power0.990.0660.18
      0.020Power0.990.0460.15
      0.040Power0.990.1300.25
      0.080Power0.990.0300.12
    • Table 5. Fitting results of the relationship between distortion and magnification of EDFA under different noise levels

      View table

      Table 5. Fitting results of the relationship between distortion and magnification of EDFA under different noise levels

      σFitting curveR2SSERMSE
      0Linear1.0000
      0.005Double exponential0.991.52×10-121.23×10-6
      0.010Double exponential0.991.76×10-114.20×10-6
      0.020Power0.966.15×10-81.75×10-4
      0.040Power0.991.82×10-73.02×10-4
      0.080Power0.995.92×10-75.44×10-4
    • Table 6. Fitting results of the relationship between SNR of vibration detection and vibration position under different noise levels

      View table

      Table 6. Fitting results of the relationship between SNR of vibration detection and vibration position under different noise levels

      σFitting curveR2SSE /dB2RMSE
      0.005linear0.994.800.77
      0.010linear0.984.250.77
      0.020linear0.991.110.43
      0.040linear0.990.470.34
      0.080linear0.980.500.50
    • Table 7. Fitting results of the relationship between distortion and vibration position under different noise levels

      View table

      Table 7. Fitting results of the relationship between distortion and vibration position under different noise levels

      σFitting curveR2SSERMSE
      0.005Power0.991.73×10-71.57×10-4
      0.010Power0.994.48×10-88.64×10-5
      0.020Power0.991.64×10-71.81×10-4
      0.040Power0.996.96×10-74.81×10-4
      0.080Power0.992.89×10-75.37×10-4
    Tools

    Get Citation

    Copy Citation Text

    Lijuan Zhao, Xuzhe Zhang, Zhiniu Xu, Yonghui Chen. Influencing Factors of IQ Demodulation Method in Distributed Acoustic Sensors[J]. Acta Optica Sinica, 2023, 43(14): 1428001

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Remote Sensing and Sensors

    Received: Feb. 6, 2023

    Accepted: Apr. 10, 2023

    Published Online: Jul. 13, 2023

    The Author Email: Zhiniu Xu (wzcnjxx@163.com)

    DOI:10.3788/AOS230508

    Topics