差分光学吸收光谱法(DOAS)已经成为测量大气中微量气体成分含量常用的方法,该方法是通过窄带分子的特征吸收波段来区分微量气体种类；并基于最小二乘原理,利用测量的大气光谱的差分吸收截面与标准的吸收截面进行拟合,确定待测气体的浓度。但在实际测量中由于系统噪声叠加在吸收光谱上,会影响测量精度。差分吸收光谱系统中惯用的方法采用多项式平滑滤波去除噪声,提出利用软阈值小波变换去噪,并对实验结果进行比较,发现软阈值小波去噪,可以提高差分吸收光谱系统的测量精度,降低差分吸收光谱系统的检测限。

Differential optical absorption spectroscopy (DOAS) has become a widely used method to measure trace gases in the atmosphere, which detects trace gases by analyzing the typical absorption waveband of narrow band molecules. The concentration of trace gases is retrieved through least-squares fits of reference absorbing cross section to the differential cross section. But in the application, for the superposition of noises on the absorption spectra, the measurement precision is affected. In the DOAS, denoising by polynomial smoothing filtering is the traditional method. Wavelet transform with soft threshold is put forward to reduce noises, which improves the precision of DOAS and declines the measuring limit, according to the comparison with experimental results.