Tunable Diode Laser Absorption Spectroscopy (TDLAS) technology has been widely used in the field of trace gas concentration detection. Within this domain, Wavelength Modulation Spectroscopy (WMS) techniques are effective in reducing the impact of low-frequency noise (1/f noise) during detection. However, noise from optical components and electronic devices can still interfere with the extracted second harmonic (2f) signal, affecting the system's accuracy in detecting trace gas concentrations. Considering that the system's noise can be approximated as a set of non-stationary signals, a study has explored an adaptive filtering algorithm based on the Least Mean Square (LMS) error to suppress the noise. This method involves extracting the reference noise signal through cross-correlation analysis and then adaptively optimizing the noisy second harmonic signal. Experimental calibration was conducted on methane of different volume concentrations and their corresponding 2f signal amplitudes using a laboratory-built TDLAS detection system. The results indicate that the proposed method can enhance the system's accuracy and stability. This research outcome could provide valuable insights for achieving high-precision detection of trace methane gas in areas such as energy extraction and environmental monitoring.
 

tunable diode laser absorption spectroscopy,wavelength modulation spectroscopy,cross-correlation analysis,adaptive filtering,second harmonic