Computational intelligence techniques to detect toxic gas presence

Abstract

The detection of toxic gas in industrial environments (performed by means of an array of low-cost on-chip chemical sensors) is a valuable approach to increase daily safety. The aim of this paper is to critically discuss the use of a-priori knowledge in the design of gas sensor systems implementing computational intelligence techniques for signal processing and gas presence detection. The availability of a-priori information about the probability density function of the considered classes as well as about the class separation boundary (Bayes boundary) allow the classifier designer for selecting appropriate condensing and editing techniques to keep under control the computational complexity