The purpose of this study is to develop a portable electronic nose system to measure volatile components of agricultural and food products. Also, operating software to control the electronic nose system through the Internet was developed. Various experiments to find optimum operating conditions of the system were performed. An array of commercial metal oxide gas sensors was used to detect various gas components. For kochujang experiment, flavour signal patterns were different to the type of kochujang. Transient and steady state signals were analyzed. Transient signal analysis was more useful in PCA. Cluster analysis showed the possibility of reduction of the number of sensors. It is feasible to optimise the kochujang making process.