An Improved Vital Signal Extraction Method Based on Laser Doppler Effect

Figures

• 

  Figure 1. Schematic diagram of the Doppler-LIDAR system. When the vibrating object is a human being, the laser beam should be aimed at the chest of the person. AOM stands for Acousto-Optic Modulator.

• 

  Figure 2. Flowchart of the algorithm.

• 

  Figure 3. Simulated respiration and heartbeat waveforms.

• 

  Figure 4. Simulated STFT plot showing IF interference.

• 

  Figure 5. STFT plot obtained after filtering the signal using filters with different pre-set frequency ranges. If the range is too narrow, residual IF interference would remain, and if the frequency range is too wide, even part of the original signal would be filtered out. Because the image intensity has been normalized, the two images have different colors.

• 

  Figure 6. CoV distribution and the time-frequency graph obtained after filtering the IF interference. In the image of CoV values, the red straight line indicates the threshold, and the blue curve indicates the CoV values.

• 

  Figure 7. Time-frequency graph containing noise and the centroid curve extracted directly from the graph.

• 

  Figure 8. Scatters present in the graph. Some of the scatters relate to valid information, while others relate to noise.

• 

  Figure 9. The graph after filtering and the centroid curve extracted from the filtered graph. The curve has a clear shape.

• 

  Figure 10. Simulatedheartbeat-andrespiratorysignalwaveforms extactedfrom the centroid curve.

• 

  Figure 11. The method to mark the hidden peaks.

• 

  Figure 12. Peaks and valleys of the heartbeat waveform. In this image, the cyan line represents the peak positions, the red line represents the valley positions, and the orange and green lines represent the extremum positions of the first-order difference of the curve.