Dynamic Response Based Odour Classification Using MOS Gas Sensors

This paper describes the classification of different sample gases based on the dynamic responses of MOS based gas sensors using artificial neural network. The dynamic responses achieved by modulating the temperature profile were used for further analysis. Principal Component Analysis (PCA) was used to visualise the different sample gas patterns . Data classification was performed using supervised neural network classifiers, namely the Multi-Layer Perceptron (MLP) network and Radial Basis Function (RBF) network and the classification accuracy for each of the two methods was determined.