There have been many developments in electronics to create real life environments with respect to sight and sound. But there are three other significant senses- smells, touch and taste which have not been experimented much with. The sensor technology of artificial olfaction had its beginnings with the invention of the first gas multi sensor array in 1982. Advances in aroma-sensor technology, electronics, biochemistry and artificial intelligence made it possible to develop devices capable of measuring and characterizing volatile aromas released from a multitude of sources for numerous applications. These devices, known as electronic noses, were engineered to mimic the mammalian olfactory system within an instrument designed to obtain repeatable measurements, allowing identifications and classifications of aroma mixtures while eliminating operator fatigue.
A few smell-sensing instruments had been proposed in narrow applications earlier in the 1960’s. Moncrief developed one of the first smell detection instruments in 1961 for agricultural application, where he used a single coated thermistor as the smell sensor. In 1964, Wilkens and Hartmen developed a smell detector where an array of smell detectors was used .It was in late 1980’s that the first intelligent electronic smelling system came into being. Researchers in the University of Warwick in Coventry, England, developed sensor arrays for odour detection. Pattern recognition techniques were used by Gardner to discriminate the output of electronic smell sensors. Hartfield described an integrated circuit based device that performs data acquisition from a miniature array of 32 conducting polymer gas sensors. David and Gardner designed a circuit capable of measuring signals from arrays of resistive and piezoelectric sensor types in the same board.
Electronic noses were originally used for quality control applications in the food, beverage and cosmetics industries. Current applications include detection of hazardous chemicals and explosives, detection of odours specific to diseases for medical diagnosis, and detection of pollutants and gas leaks for environmental protection. The Warwick pioneers envisioned an electronic equivalent of the mammalian olfactory system and so even though it doesn’t resemble its biological counterpart the least bit, the label ‘electronic nose’ or ‘E nose’ has been widely adopted.
The Electronic Nose Overview
The electronic nose is a system that consists of three functional components that operate serially on an odorant sample-a sample handler, an array of gas sensors and a signal processing system. The output of the electronic nose can be the identity of the odorant, an estimate of the concentration of the odorant, or the characteristics properties of the odour as might be perceived by a human.
ENS can be seen as arrays of-non specific sensors able to generate electrical signals in response to either simple or complex volatile compounds, and give through a suitable multi component analysis technique, the possibility of discrimination, recognition and classification of odours. The target compound, in gaseous form, is introduced into the sensing chamber where the sensors are exposed to the vapour. A variety of basic sensors can be used according to the nose strategy chosen. Some of them are sensitive to the mass of adsorbed species, others show sensitivities to electric charges while others are based on either surface or bulk conductivity changes due to chemically interactive materials. These changes are dependent on a complex interaction between the components of the vapour and the sensors, as each sensor responds to a number of components in a unique manner. Each sensor in the array has different characteristics (e.g., coatings, operating temperatures, etc.) and, hence, each sensor will give a different electrical response (voltage output) for a particular odour. The pattern of response across all the sensors in the array is used to identify and/or characterise the odour.
In electronic noses pattern recognition methods are required for the qualitative analysis of odours or of different compounds present in a certain mixture and multi component analysis methods are required for the quantitative determination of one or more compounds in a mixture. Commercially available analysis techniques fall into three main categories as follows :
1. Graphical analyses: bar chart, profile, polar and offset polar plots
2. Multivariate data analyses (MDA): principal component analysis (PCA)
3. Network analyses: artificial neural network (ANN) and radial basis function (RBF) Experimental data are evaluated by a qualitative or quantitative link between output signals of an instrument and the chemical information (composition or concentration of analytes). This requires a comparison of the sensor outputs with previously recorded calibration data. When high concentrations of volatile are measured, a non-linear pattern recognition technique, such as ANN (Artificial Neural Networks), would be more appropriate . On a very simplified and abstract level, ANN is based on the cognitive process of the human brain. ANNs are a commonly used pattern recognition technique which attempt to mimic the biological processes of the human brain.
An “electronic nose” is a system originally created to mimic the function of an animal nose. However, this analytical instrument is more a “multi-sensor array technology” than a real “nose”. Whatever the sensor technology, it is still far from the sensitivity and selectivity of a mammalian nose. Therefore, its aim is not to totally replace either the human nose or other analytical methods. A sensory panel is necessary to define the desired product quality which can then be used to train the system. Traditional analytical methods such as GC-analysis will always be needed to determine qualitatively or/and quantitatively why one food sample differs from others. The “electronic nose” can only perform quick “yes or no” tests in comparison to other products. It could occasionally replace sensory analysis and even perform better than a sensory panel in routine work, or in cases where non-odorous or irritant gases need to be detected.
Compared to classical and other novel analytical methods, the electronic nose built offers a cheap and non destructive instrument that (if properly programmed and automated) can be operated by non specialists. The number of measurements that can be done in a day compares favourably to other sophisticated methods, such as aromatic profile identification using chromatography (one of the newest approaches), and since the whole process is automatic, the cost of each measurement is very low. Therefore, in the near future, the electronic nose could be envisaged as a global measurement system calibrated for ripeness determination or a multi instrument system to extract the indicators for which it has been calibrated.Food analysis is a very complex discipline. Due to its strict interaction with the quality of life it is extremely important to improve the performances of the methods in the fields. EN seems to be a new instrument that can offer the unique advantage of providing fast and low expensive qualitative analysis of many kinds of foods.
Further work needs to address important limitations. For example, a straightforward procedure should be devised to detect and correct sensor drift from year to year. Also, the initial calibration of the system for a given cultivar should take only a few measurements and be valid, at least, for some consecutive campaigns. Finally, the measurement cycle should be faster in order to increase throughput. All of these considerations are being studied and might imply the optimization of the sampling process, the use of more advanced processing algorithms, and the incorporation of new sensor technologies into the system.