Strategies for Mimicking Olfaction: The Next Generation of Electronic Noses?

Pearce, Timothy Charles; Gardner, Julian W.; Göpel, Wolfgang

doi:10.1002/1616-8984(199801)3:1<61::aid-seup61>3.0.co;2-7

Cited by 7 publications

(5 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Global methods, on the other hand, are typically used to ensure that sensor magnitudes are comparable, preventing subsequent pattern-recognition procedures from being overwhelmed by sensors with arbitrarily large values. Two global procedures are commonly employed in e-nose systems: (i) sensor autoscaling, [5]. Table I provides a few additional references on signal preprocessing from the e-nose literature.…”

Section: Preprocessing Techniques For Gas Sensor Arrayssupporting

confidence: 46%

“…Ridge regression (RR) is a regularization method that stabilizes the OLS solution by adding a multiple of the identity matrix 5 to the estimation of the covariance matrix [48], [69] where is a regularization parameter that controls the amount of shrinkage toward the identity matrix. For , RR is equivalent to OLS, whereas for , the solution becomes a constant model that always predicts the mean concentration of the training data.…”

Section: B Ridge Regressionmentioning

confidence: 47%

“…However, PCR presents similar problems to PCA since the directions of maximum variance in may not necessarily be correlated with the dependent variables. 5 This expression assumes that the sensors have the same variance. Otherwise, the regularization term tr(Y Y )I=M must be replaced by a diagonal matrix of the sensor variances [70].…”

Section: Principal Components Regressionmentioning

confidence: 49%

See 2 more Smart Citations

Pattern analysis for machine olfaction: a review

2002

View full text Add to dashboard Cite

Abstract-Pattern analysis constitutes a critical building block in the development of gas sensor array instruments capable of detecting, identifying, and measuring volatile compounds, a technology that has been proposed as an artificial substitute of the human olfactory system. The successful design of a pattern analysis system for machine olfaction requires a careful consideration of the various issues involved in processing multivariate data: signal-preprocessing, feature extraction, feature selection, classification, regression, clustering, and validation. A considerable number of methods from statistical pattern recognition, neural networks, chemometrics, machine learning, and biological cybernetics has been used to process electronic nose data. The objective of this review paper is to provide a summary and guidelines for using the most widely used pattern analysis techniques, as well as to identify research directions that are at the frontier of sensor-based machine olfaction.

show abstract

Section: Preprocessing Techniques For Gas Sensor Arrayssupporting

confidence: 46%

Section: B Ridge Regressionmentioning

confidence: 47%

Section: Principal Components Regressionmentioning

confidence: 49%

See 1 more Smart Citation

Pattern analysis for machine olfaction: a review

2002

View full text Add to dashboard Cite

show abstract

“…In case of handling these data, a method using the useful parts of the data can be effective but requires various analysis techniques, generally principal components analysis (PCA), Fisher's linear discriminant analysis (LDA), feature subset selection (FSS), C-means clustering, radial basis function classification, K-nearest neighbor classification and so on [1][2][3][4][5]. To apply the above techniques, in addition, complex or various numerical analyses are accompanied independently [6][7][8][9][10]. Recently, soft computing techniques such as fuzzy logic, neural networks and genetic algorithms have been frequently used as the strong tools to handle the uncertain data.…”

Section: Introductionsupporting

confidence: 40%

Effective Gas Identification Model based on Fuzzy Logic and Hybrid Genetic Algorithms

Bang¹,

Byun

Lee³

2012

Journal of Sensor Science and Technology

View full text Add to dashboard Cite

This paper presents an effective design method for a gas identification system. The design method adopted the sequential combination between the hybrid genetic algorithms and the TSK fuzzy logic system. First, the sensor grouping method by hybrid genetic algorithms led the effective dimensional reduction as well as effective pattern analysis from a large volume of pattern dimensions. Second, the fuzzy identification sub-models allowed handling the uncertainty of the sensor data extensively. By these advantages, the proposed identification model demonstrated high accuracy rates for identifying the five different types of gases; it was confirmed throughout the experimental trials.

show abstract

“…The electronic nose is in a way similar to the human olfactory sense [7,8]. The concept of the electronic nose system emerged from progress in a number of key research areas.…”

Section: Electronic Nose/electronic Tonguementioning

confidence: 99%

Selected examples of intelligent (micro) sensor systems: state-of-the-art and tendencies

Hauptmann

2006

Meas. Sci. Technol.

View full text Add to dashboard Cite

The capability of intelligent sensors to have more intelligence built into them continues to drive their application in areas including automotive, aerospace and defense, industrial, intelligent house and wear, medical and homeland security. In principle it is difficult to overestimate the importance of intelligent (micro) sensors or sensor systems within advanced societies but one characteristic feature is the global market for sensors, which is now about $20 billion annually. Therefore sensors or sensor systems play a dominant role in many fields from the macro sensor in manufacturing industry down to the miniaturized sensor for medical applications. The diversity of sensors precludes a complete description of the state-of-the-art; selected examples will illustrate the current situation. MEMS (microelectromechanical systems) devices are of special interest in the context of micro sensor systems. In past the main requirements of a sensor were in terms of metrological performance. The electrical (or optical) signal produced by the sensor needed to match the measure relatively accurately. Such basic functionality is no longer sufficient. Data processing near the sensor, the extraction of more information than just the direct sensor information by signal analysis, system aspects and multi-sensor information are the new demands. A shifting can be observed away from aiming to design perfect single-function transducers and towards the utilization of system-based sensors as system components. In the ideal case such systems contain sensors, actuators and electronics. They can be realized in monolithic, hybrid or discrete form—which kind is used depends on the application. In this article the state-of-the-art of intelligent sensors or sensor systems is reviewed using selected examples. Future trends are deduced.

show abstract

Strategies for Mimicking Olfaction: The Next Generation of Electronic Noses?

Cited by 7 publications

References 46 publications

Pattern analysis for machine olfaction: a review

Pattern analysis for machine olfaction: a review

Effective Gas Identification Model based on Fuzzy Logic and Hybrid Genetic Algorithms

Selected examples of intelligent (micro) sensor systems: state-of-the-art and tendencies

Contact Info

Product

Resources

About