The optimum embedded controller for handheld electronic nose

Abstract

Electronic nose (e-nose) is a non-destructive intelligent instrument that mimics human olfactory system to detect, discriminate and classify odour. The instrument have vast potential applications includes food quality assurance, plant disease and malodour monitoring. The increases of the instrument potential applications have attracted many research groups to developed a cost-effective system with simple operating procedure. Recent developments in embedded technology have made possible for low cost integration of powerful embedded system for a small device. This paper discusses the selection of optimum embedded controller for the development of a handheld e-nose. The selected controller should enable the instrument to operate effectively. The developed instrument is using off-theshelf components i.e. metal oxide sensor, microcontroller and signal conditioning circuit. The instrument offer rapid response, versatility and novelty in the detection of sample odour. The data processing is using multivariate statistical analysis i.e. principal component analysis (PCA), Hierarchical Cluster Analysis (HCA) and Linear Discriminate Analysis (LDA). The developed instrument is tested to discriminate the basic aromatic smell. Initial results show that the instrument is able to discriminate the samples based on their odour chemical fingerprint profile. The multivariate statistical analysis (PCA, HCA and LDA) plot show that the samples are grouping into different cluster.