Multicomponent gas mixture analysis using a single tin oxide sensor and dynamic pattern recognition

Abstract

A new method, which is based on the discrete wavelet transform, is presented for extracting important features from the response transients of a micromachined, tin oxide-based gas sensor. It is shown that two components in a mixture can be simultaneously and accurately quantified by processing the response dynamics of a single sensor operated in a temperature-modulated mode. The discrete wavelet transform outperforms the fast Fourier transform (classical approach) because it is more appropriate for the non- linear frequency-time problem encountered here.