An Adaptive Hyperfine Spectrum Extraction Algorithm for Optical Sensing Based on SG Filtering and VMD.

Area of Science:

• Optical Sensing
• Spectroscopy
• Signal Processing

Background:

• Signal demodulation in optical sensing often degrades spectral data due to noise like Doppler and aliasing.
• Existing algorithms struggle with complex spectral signals or require extensive parameter tuning.
• Effective noise suppression and feature preservation in optical spectroscopy remain challenging.

Purpose of the Study:

• To develop an adaptive spectral extraction algorithm for improved optical sensing measurements.
• To address limitations of existing methods in handling complex spectral data and noise.
• To enhance spectral signal quality and reliability in optical instrumentation.

Main Methods:

• Proposed an adaptive spectral extraction algorithm integrating Variational Mode Decomposition (VMD) and Savitzky-Golay (SG) filtering.
• Implemented an innovative adaptation strategy to optimize VMD mode number and SG filter parameters (frame length, order).
• Leveraged time-domain and frequency spectrum information for adaptive parameter determination.

Main Results:

• The adaptive VMD-SG algorithm effectively suppresses noise while preserving crucial spectral features.
• Validated through simulations and experiments, demonstrating significant enhancement in spectral signal quality.
• Successfully restored absorption peaks and eliminated the need for manual parameter adjustments.

Conclusions:

• The proposed adaptive spectral extraction method offers a robust solution for optical sensing.
• Significantly improves measurement accuracy and reliability, especially for complex spectral analyses.
• Provides a valuable tool for advancing optical sensing instrumentation and applications.