Gas Detection Using Resistive Gas Sensors And Radial Basis Function Neural Networks

We present a use of Radial Basis Function (RBF) neural networks and Fluctuation Enhanced Sensing (FES) method in gas detection system utilizing a prototype resistive WO3 gas sensing layer with gold nanoparticles. We investigated accuracy of gas detection for three different preprocessing methods: no preprocessing, Principal Component Analysis (PCA) and wavelet transformation. Low frequency noise voltage observed in resistive gas sensor was treated as input data of preprocessing methods. The power spectral density was computed for two firstly enumerated methods to improve effectiveness of gas detection. The PCA method preserves the most informative part of power spectral density by reducing size of input data and gave slightly worse results. The best results secured wavelet transform. We have compared the reported results with our previous work about Least Squares Support Vector Machines (LS-SVM) algorithm. We conclude that the applied method is much simpler and faster than the previous one and secured similar gas detection accuracy.