Wavelet transform-based fast feature extraction from temperature modulated semiconductor gas sensors

We demonstrate that a single, thermally modulated tungsten oxide-based resistive sensor can discriminate between different vapours. The method uses a novel feature extraction and pattern classification method, which is based on the discrete wavelet transform (DWT). It was found that DWT outperformed fast Fourier transform (FFT) in the extraction of important features from the sensor response and, allowed for straightforward gas recognition in feature space.