Multi-class system based on SVM for real-time gas mixture classification

In this work we address the use of support vector machine (SVM) in the multi-class gas classification system. In our case, the objective is to classify single gases and their gas mixture with a semiconductor-type electronic nose. The support vector machine is a powerful tool for solving the problem with small sampling, non-linearity and high dimension. The SVM has some typical multi-class classification models, which are One vs One (OVO) and One vs All (OVA). But their studies show weaknesses on calculation time, decision time and the reject region. We propose a method of multi class system based on the SVM using hierarchical clustering analysis for real time gas classification. Experimental results show that the proposed method has the best performance compared with the typical multi-class systems based on the SVM, and this method make it possible to classify easily and fast in the real embedded system compared with BP-MLP and Fuzzy ARTMAP.