Wavelet Denoising in Near-Infrared Broadband Cavity-Enhanced Absorption Spectroscopy

In order to effectively suppress the noise in gas detection and improve the inversion precision of gas concentration, we investigate a wavelet denoising algorithm for near-infrared broadband cavity-enhanced gas sensing system. Optimization analysis of wavelet denoising shows that it can achieve the optimal denoising effect by using db2 wavelet function as the wavelet base to perform 6-layer denoising on the polluted signal, and at the same time, by using heursure threshold estimation method and local threshold to zero the noise part wavelet coefficient. The near-infrared broadband cavity-enhanced absorption spectroscopy technique combined with a high-resolution Fourier transform infrared spectrometer is used to establish a gas sensing system for methane detection. The concentration inversion of methane absorption coefficient before and after wavelet denoising is performed using a least square fitting algorithm. Experimental results show that the inversed concentration results with wavelet denoising arc closer to the true value than those without denoising. The inversion accuracy is improved by 7%, the signal-to-noise ratio is increased by 90%, and the system detection limit is reduced by 45%. It is evidenced that the wavelet denoising algorithm can effectively improve the detection accuracy.