Photoplethysmography (PPG) is a widely used optical technique for measuring blood volume changes in the microvascular bed of tissue.
The effectiveness of PPG for biometric applications is significantly influenced by the wavelength of light used for signal acquisition. Different wavelengths interact with skin and blood in distinct ways, affecting the quality and characteristics of the PPG signal.
Wavelength Characteristics and Signal Quality
1. Penetration Depth:
– Different wavelengths penetrate skin tissues at varying depths. For instance, longer wavelengths (e.g., infrared at 940 nm) can penetrate deeper than shorter wavelengths (e.g., green light at 525 nm). This characteristic is crucial for capturing PPG signals from deeper blood vessels, which can enhance the reliability of measurements, especially in individuals with different skin types and thicknesses[5][9].
2. Absorption by Hemoglobin and Melanin:
– The absorption spectra of hemoglobin and melanin vary with wavelength. Shorter wavelengths, such as blue and green light, are absorbed more by melanin, making them less effective for individuals with darker skin tones. Conversely, infrared light is less absorbed by melanin, allowing for better signal acquisition in these populations[4][9]. Studies have shown that green light tends to provide a higher signal-to-noise ratio (SNR) under resting conditions, while blue and green wavelengths perform better during physical activity due to increased blood volume changes[4].
3. Signal Modulation:
– The modulation of the PPG signal, which reflects changes in blood volume with each heartbeat, is also affected by wavelength. Research indicates that green light provides larger mean modulation compared to red or infrared light, particularly in resting conditions. This modulation is crucial for accurate biometric identification, as it directly relates to the quality of the PPG waveform[4][9].
Optimal Wavelength Selection
1. Skin Tone Considerations:
– The choice of wavelength should consider the skin tone of the individual being measured. For lighter skin tones, shorter wavelengths may yield better results, while for darker skin tones, longer wavelengths may be preferable to minimize absorption by melanin[2][4].
2. Multi-Wavelength Approaches:
– Using multiple wavelengths can enhance the robustness of PPG signal acquisition. Multi-wavelength PPG (MW-PPG) systems can combine signals from different wavelengths to improve SNR and reduce the effects of noise and artifacts. Such systems have shown promise in providing more stable and accurate PPG signals across diverse populations and conditions[8][9].
3. Device Design and Sensor Layout:
– The arrangement of light sources and photodetectors also plays a critical role in PPG signal quality. Optimizing the source/detector separation and ensuring proper alignment can significantly impact the performance of PPG devices, particularly when using different wavelengths[4][9].
The selection of light wavelengths for PPG signal acquisition is a critical factor in biometric applications. Understanding the interactions between light, skin, and blood at various wavelengths can lead to improved biometric systems that are more accurate and reliable across different populations. Future developments in PPG technology should focus on optimizing wavelength selection and sensor design to enhance the fidelity of biometric identification and authentication.
Citations:
[1] https://www.researchgate.net/figure/PPG-signal-for-different-wavelengths-with-standardization-of-LED-power_fig5_355117838
[2] https://www.researchgate.net/figure/Mean-PPG-signal-shapes-for-each-wavelength-with-standard-deviation-SD_fig3_253375174
[3] https://trepo.tuni.fi/bitstream/10024/154951/5/AdeyemiAkinlabi.pdf
[4] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8073123/
[5] https://www.researchgate.net/publication/23932320_Comparison_of_Reflected_Green_Light_and_Infrared_Photoplethysmography
[6] https://www.researchgate.net/publication/322049222_Evaluation_of_PPG_Biometrics_for_Authentication_in_different_states
[7] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4848078/
[8] https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7698030/
[9] https://www.frontiersin.org/journals/physiology/articles/10.3389/fphys.2021.808451/full
English 
Հայերեն 
Polski