A matched-profile method for simple and robust vapor recognition in electronic nose (E-nose) system

Abstract

The vapor recognition performance of electronic nose (E-nose) system can be improved by manipulating the sensor array responses of vapors in profile forms. The sensor array composed of various carbon-black (CB) polymer composites shows featured response profile patterns, varied from vapors to vapors due to different chemical interactions between the arrayed materials and the vapors. These multidimensional sensory data gives more information than collection of the piecemeal signal features, i.e., maximum sensitivity, signal slopes, rising-time. To use them in vapor recognition task, we proposed a novel matched-profile method was proposed, which is based on the typical digital image pattern matching. The degrees of matching between eight different vapors were evaluated by using the proposed method. The vapor responses are measured by the silicon-based gas sensor array with 16 CB polymer composites installed in membrane structure. The results showed higher contrast between matching and non-matching vapors than conventional method. This implies reliable vapor recognition in E-nose system.