Temperature modulation and artificial neural network evaluation for improving the CO selectivity of SnO2 gas sensor

Stannic oxide sensors were developed to monitor CO concentrations 10-250 ppm. Cross sensitivities of these sensors against methane 100-2000 ppm can be suppressed by evaluating the features extracted from the sensor signals. For this purpose, the working temperature of the sensor was modulated between 250°C and 300°C and the dynamic responses to different concentrations of CO, CH4, and their mixtures were measured. The discrete wavelet transform (DWT) was used to extract important features from the sensor response. These features were then input to pattern recognition (neural) method. The species considered can be discriminated with a 100% success rate using back propagation network and the concentrations of the gases studied can also be accurately predicted.