A near-infrared methane (CH4) gas sensor system was implemented using a quartz crystal tuning fork (QCTF) based photoelectric detector and a fiber-coupled distributed feedback diode laser emitting at 1.653 μm. Wavelength-modulation with second-harmonic (2f) detection method was investigated. For signal processing, two different demodulation algorithms, i.e. Fast Fourier Transformation (FFT) and digital lock-in amplifier (DLIA), were developed and completely compared for harmonic signal extraction. Experimental results show that the SNR of WMS-2f signals using the proposed FFT algorithm can be improved by 2.5 times on average than the common DLIA method, which agrees well with the theoretical value of 2.7 times achieved with the simulated spectra data. In comparison, the proposed FFT algorithm shows better performance on signal processing efficiency and simplification, when a QCTF was used as a photoelectric detector or an acoustic transducer.