Wireless electronic nose system for real-time quantitative analysis of gas mixtures using micro-gas sensor array and neuro-fuzzy network

Abstract

A wireless electronic nose system (WENS) is designed for the real-time quantification of ammonia (NH3), hydrogen sulfide (H2S), and their mixtures. The WENS hardware consists of a microcontroller for obtaining measurement data from a micro-gas sensor array, and an RF transceiver for transmitting the data sets to a master sensor node. Meanwhile, the WENS software analyses the binary gas mixtures using a fuzzy ARTMAP classifier and a fuzzy ART-based concentration estimator with multiplicative drift correction based on reference gases. A virtual instrument is developed in the LabVIEW environment for monitoring the analyzed gas mixtures. The performance of the proposed WENS is also assessed and compared with the minimum and product inference methods. The proposed WENS adopting the weighted inference method produces the best concentration estimations as regards the root mean square error. © 2008 Elsevier B.V. All rights reserved.