Enhanced Sensitivity of CO Photoacoustic Sensors Using Empirical Mode Decomposition Denoising Algorithm

A CO sensor based on photoacoustic spectroscopy (PAS) with empirical mode decomposition (EMD) algorithm is investigated and demonstrated in this paper. In the PAS system, the complicated photo-thermal-acoustic conversion is a nonlinear and non-stationary process and contains various noise. In order to compensate the low signal-to-noise ratio (SNR), the EMD is introduced in the PAS system to deal with the photoacoustic signal. The experimental results show that a gain factor of ∼3.0 on the SNR is achieved and the sensor has an excellent linear response to the gas concentration. The minimum detection level (MDL) for CO detection is reduced to 1.14 ppm with a 300 ms integrated time at room temperature and atmospheric pressure.