ABSTRACT
Due to recent advances in different interdisciplinary matters, such as organic and inorganic chemistry, sensor technology, electronics and artificial intelligence, the fabrication of electronic noses for the characterisation of aromas has been possible and it has become a commercial reality [1,2]. Along this decade this technology will find a wide range of commercial applications, such as medical diagnostics, environmental monitoring, and processing and quality control of foods. Within the former application it is the aroma of Iberian dry-cured ham, which is one of the most important characteristics that influences the acceptance of this product by consumers. The study of dry-cured ham flavour is very interesting to understand the formation of odorous compounds during the curing process. The knowledge of these compounds in each stage is very important for the food industry so that the process to obtain the final product can be controlled. Until now, there have been few studies on the aroma components of hams and most studies have been camed out with sophisticated and expensive instrumentation such as gas-phase chromatography-mass spectrometry and extraction techniques such as high vacuum distillation and dynamic headspace. Such procedures have allowed to identify seventy-seven components in the ham flavour, most of which were alkanes, aldehydes, aliphatic alcohols, ketones and esters [3-51. The great shortcoming of these techniques is that they are not able to measure the compounds in real time and in continuous, mainly during the stages of elaboration of the product. Furthermore, analysis time is very long, so that they can not be used routinely in the food industry. In this paper we present an electronic nose (chemical sensor array) which is able to perform measurements in real time, in continuous, and in a fast way, so that direct analysis of complex gaseous mixtures (aroma) can be made [6,7]. The response of each array sensor is related to the whole of the species present in the environment of measurement due to its low selectivity. Therefore the sensor array system uses the global information formed for all the responses to classify the environment, and it extracts the information based on the requirement that each sensor must have different response values. In our case the obtained signal from each sensor is transmitted directly via Internet to anywhere to be evaluated later by pattern recognition methods, so that selectivity and discrimination capability can be improved. Thus, we can obtain very rapid diagnoses for the quality control of the product.
