A thin-film SnO2 sensor system for simultaneous detection of CO and NO2 with neural signal evaluation

Simultaneous CO and NO2 measurements are of importance for the ventilation control of automobiles and other applications, For this purpose semiconducting SnO2 sensors are often used. A well known disadvantage of SnO2 sensors is the concurrent reaction of the oxidizing NO2 and the reducing CO on the sensor surface, which causes a near zero sensor signal in the presence of both gases in a certain range of mixtures. A second disadvantage of SnO2 sensors are the long rise and decay times of the sensor signal. The combination of different SnO2 sensors, operated at different temperatures and combined with a signal evaluation system based on a specially trained neural forward network (artificial neural net (ANN)) solves this problem. The runtime version of the neural net is a small program, compatible with customary micro controllers. These signal evaluation techniques are applicable to similar problems using sensor arrays or single sensors in a non-stationary operating mode.