Wavelet transform smoothing filters for metal oxide gas sensor signal cleaning

Abstract

This paper reports on a series of experiments to evaluate the methods for feature extraction and denoising the digital signal from thin film zinc oxide-tin dioxide composite gas sensor devices. The aim was to find a method that not only cleaned the signal but also maintained the shape, precision and resolution of the signal. It was found that the Savitzky-Golay smoothing filter method gave the best, smoothed and cleaned, approximation of the sensor response regardless of the thin film composition, target gas concentration or operating temperature.