Feature Domain Transform Filter for the Removal of Inherent Noise Bound to the Absorption Signal

Abstract

We propose to replace the traditional time-frequency domain filtering with feature domain filtering to realize an innovation of filtering algorithm. A feature domain transform filter (FDTF) is composed of the feature domain transform layer based on principal component analysis (PCA) algorithm, the feature domain information extractor based on deep learning and the time domain transform layer. It is established to filter out the noise with the same frequency and phase as the signal and is verified on methane gas. Although FDTF is established based on the simulated data set, the filtering effects of the simulation test set and the experimental data set show that the proposed FDTF outperforms other widely used time-frequency filtering algorithms. The FDTF-assisted methane sensor has good linearity at different concentrations of methane gas. With the FDTF enhancement, the optimized methane sensor performs excellent precision and stability in real-time measurements and achieves the minimum detectable column density of 2.50 ppm• m. This is undoubtedly a successful attempt to move the signal to a new domain for parsing and separation.

Research on VMD-Based Adaptive TDLAS Signal Denoising Method

“…The laser output's optical frequency and optical power after the scanning signal and modulation signal are injected into the laser can be expressed as [5]…”

“…According to the optimal solution (K and α) obtained in step (5), use VMD to decompose, obtain K IMF components, and calculate each IMF component with the standard signal correlation coefficient; 7.…”

“…Repeat steps (3) to ( 5) until the given number of iterations is reached and the op solution of the particle is output; 6. According to the optimal solution (K and α) obtained in step (5), use VMD to de pose, obtain K IMF components, and calculate each IMF component with the st ard signal correlation coefficient; 7. According to the given threshold value of the correlation coefficient, choose the components to reconstruct the signal and add the selected IMF components to o the restructured signal after the noise is reduced.…”

“…There are two general ways to improve the sensitivity and SNR of a signal in a TDLAS system: one is to enhance the signal, and the other is to reduce the noise [4]. However, the methods of signal enhancement usually change the system's structure, which leads to an increase in cost [5]. The way to reduce noise in TDLAS systems usually uses software filtering [6,7], which only needs partial computational resources and does not require additional electronic and optical devices [8][9][10][11].…”