What is Electronic Nose ?

        The Electronic Nose utilizes the concepts of sensor technology, biochemistry, electronics and Artificial Intelligence to design the biological olfactory system with an instrument capable of identifying the aroma mixture.Typically Electronic Nose consists of a multi-sensor array, an information processing unit utilizing the concepts of Artificial Neural Network (ANN), software with digital pattern-recognition algorithms and reference Library system. The e-nose has been a great help in performing the procedures in few minutes that requires many hours or even days to complete and also in detecting the complex odors. It is a smart device that is designed to detect and discriminate among various odors using the array of sensors developed. In order to perform these kinds of odor detection, it includes the sampling system, an array of chemical gas sensors, an Analog to Digital converter, a microprocessor (based on pattern recognition system, classification system) and a controlling system.

Electronic Nose Working Principle

The electronic nose was developed in order to mimic human olfaction whose functions are non separate mechanism , i.e. the smell or flavor is perceived as a global finger print. Essentially the instrument consists of sensor array, pattern reorganization modules, and headspace sampling, to generate signal pattern that are used for characterizing smells. The electronic nose consists of three major parts which are detecting system, computing system, sample delivery system.

The sample delivery system: The sample delivery system enables the generation of headspace of sample or volatile compounds which is a fraction analyzed. The system then sends this head space into the detection system of the electronic nose.
The detection system: The detection system which consists of a group of sensors is the reactive part of the instrument. When in contact with volatile compounds at that time the sensors reacts causing changes in electrical characteristics.
The Computing system: In most electronic noses each sensor is sensitive to all molecules in their specific way. However in bioelectric noses the receptor proteins which respond to specific smell molecules are used. Most of electronic noses use sensor arrays that react to volatile compounds. Whenever the sensors sense any smell , a specific response is recorded that signal is transmitted into the digital value.

The more commonly used sensors in electronic nose are

  • Metal Oxide semiconductor (MOSFET)
  • Conducting polymers
  • Quartz crystal microbalance
  • Piezoelectric sensors
  • Metal Oxide sensors

Metal Oxide semiconductor sensor:
This is used for switching or amplifying electronic signals. The Working principle of MOSFET is that molecules entering into the sensor area will be charged positively or negatively which have directly effect on the electric field inside MOSFET.
Metal Oxide sensors: (MOS)
This sensor is based on adsorption of gas molecules to provoke change in conductivity. This conductivity change is the measure of the amount of volatile organic compounds adsorbed.
Piezoelectric sensors:
The adsorption of gas onto the surface of the polymer leads to change in mass on the sensor surface. This is turn produce a change in the resonant frequency of the crystal.

Quartz crystal microbalance:
This is a way of measuring mass per unit area by measuring the change in frequency of crystal resonator. This can be stored in a data base.
Conducting polymers:
Conductive polymer gas sensors operate based on changed in electrical resistance caused by adsorption of gases onto the sensor surface.

Data Analysis for Electronic Nose:

The digital output generated by electronic nose sensors has to be analyzed and interpreted in order to provide. There are three main types of commercially available techniques.

  • Graphical analysis
  • Multivariate data analysis
  • Network analysis

The choice of method utilized depends on available input data from sensors. The simplest form of a data reduction is a graphical analysis useful for comparing samples or comparing smells identification elements of unknown analysts relative to those of known sources in reference libraries. The multivariate data analysis generates a set of techniques for the analysis of data that is trained or untrained technique. The untrained techniques are used when a data base of known samples has not been built previously. The simplest and most widely used untrained MDA technique is a principle component analysis. The electronic nose data analysis MDA is a very useful when sensors have partially coverage sensitivities to individual compounds present in a sample mixer. The PCA is a most useful when no known sample is available. The neural network is the best known and most derived analysis techniques utilized in a statistical software packages for commercially available electronic nose.

For examples electronic nose system for the fruit smell detection:

Application of Electronic nose:

  • Medical diagnostics and health monitoring
  • Environmental monitoring
  • Application in food industry
  • Detection of explosive
  • Space applications (NASA)
  • Research and development industries
  • Quality control laboratories
  • The process and production department
  • Detection of drug smells
  • Detection of harmful bacteria

Author : Srinivasa Rao Polisetty
About The Author :  Srinivasa Rao is a iOS developer loves to design Applications and bug fixes. He is very smart, cool and passionate about new things. He is responsible in taking new challenges.
Source : Wikipedia