### Abstract

Finding a rapid gas identification algorithm with high accuracy and a closed-form solution that does not require any manual tuning of parameters is the major challenge to overcome in adopting electronic nose technology in daily life applications. Recently, bio-inspired rank-order-based classifiers have been proposed to meet this challenge by transforming multidimensional sensitivity vectors into temporally ordered spike sequences for target gases. The performance of these classifiers, however, is limited when the spike sequences corresponding to all the target gases do not contain sufficient discriminatory information to identify them. Moreover, their identification decision is delayed up to the computation of the sensitivity vectors at steady state, which incurs a long waiting time. In this paper, we adopt a Bayesian parametric method with a normal distribution model as an alternative approach that provides a closed-form solution with only second-order statistics, i.e., mean and covariance. However, for electronic nose applications, a reliable estimation of the covariance matrix is a major challenge with the commonly used maximum likelihood estimate in Bayesian inference because of the limited number of available measurements for each target gas. We exploit random matrix theory principles to reduce randomness in the sample covariance matrix for its reliable estimation. Moreover, transient features are used to accelerate the gas identification. In order to validate the effectiveness of this approach, data of eight gases, namely, C< ±0.5-ppm frequency stability_{3< ±0.5-ppm frequency stability}H< ±0.5-ppm frequency stability_{8< ±0.5-ppm frequency stability}, C< ±0.5-ppm frequency stability_{6< ±0.5-ppm frequency stability}H< ±0.5-ppm frequency stability_{6< ±0.5-ppm frequency stability}, CH< ±0.5-ppm frequency stability_{2< ±0.5-ppm frequency stability}O, CL< ±0.5-ppm frequency stability_{2< ±0.5-ppm frequency stability}, CO, CO< ±0.5-ppm frequency stability_{2< ±0.5-ppm frequency stability}, NO< ±0.5-ppm frequency stability_{2< ±0.5-ppm frequency stability}, and SO< ±0.5-ppm frequency stability_{2< ±0.5-ppm frequency stability}, are acquired in the laboratory. We achieve a 7.73% performance improvement as compared with Bayesian inference using the maximum likelihood estimate, and an overall accuracy rate of 99.40% on the experimental data set.

Original language | English |
---|---|

Article number | 7352294 |

Pages (from-to) | 2036-2045 |

Number of pages | 10 |

Journal | IEEE Sensors Journal |

Volume | 16 |

Issue number | 7 |

DOIs | |

Publication status | Published - 1 Apr 2016 |

### Fingerprint

### Keywords

- Bayesian inference
- electronic nose
- gas identification
- random matrix theory

### ASJC Scopus subject areas

- Instrumentation
- Electrical and Electronic Engineering

### Cite this

*IEEE Sensors Journal*,

*16*(7), 2036-2045. [7352294]. https://doi.org/10.1109/JSEN.2015.2507580

**Robust Bayesian Inference for Gas Identification in Electronic Nose Applications by Using Random Matrix Theory.** / Hassan, Muhammad; Bermak, Amine.

Research output: Contribution to journal › Article

*IEEE Sensors Journal*, vol. 16, no. 7, 7352294, pp. 2036-2045. https://doi.org/10.1109/JSEN.2015.2507580

}

TY - JOUR

T1 - Robust Bayesian Inference for Gas Identification in Electronic Nose Applications by Using Random Matrix Theory

AU - Hassan, Muhammad

AU - Bermak, Amine

PY - 2016/4/1

Y1 - 2016/4/1

N2 - Finding a rapid gas identification algorithm with high accuracy and a closed-form solution that does not require any manual tuning of parameters is the major challenge to overcome in adopting electronic nose technology in daily life applications. Recently, bio-inspired rank-order-based classifiers have been proposed to meet this challenge by transforming multidimensional sensitivity vectors into temporally ordered spike sequences for target gases. The performance of these classifiers, however, is limited when the spike sequences corresponding to all the target gases do not contain sufficient discriminatory information to identify them. Moreover, their identification decision is delayed up to the computation of the sensitivity vectors at steady state, which incurs a long waiting time. In this paper, we adopt a Bayesian parametric method with a normal distribution model as an alternative approach that provides a closed-form solution with only second-order statistics, i.e., mean and covariance. However, for electronic nose applications, a reliable estimation of the covariance matrix is a major challenge with the commonly used maximum likelihood estimate in Bayesian inference because of the limited number of available measurements for each target gas. We exploit random matrix theory principles to reduce randomness in the sample covariance matrix for its reliable estimation. Moreover, transient features are used to accelerate the gas identification. In order to validate the effectiveness of this approach, data of eight gases, namely, C< ±0.5-ppm frequency stability3< ±0.5-ppm frequency stabilityH< ±0.5-ppm frequency stability8< ±0.5-ppm frequency stability, C< ±0.5-ppm frequency stability6< ±0.5-ppm frequency stabilityH< ±0.5-ppm frequency stability6< ±0.5-ppm frequency stability, CH< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stabilityO, CL< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stability, CO, CO< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stability, NO< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stability, and SO< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stability, are acquired in the laboratory. We achieve a 7.73% performance improvement as compared with Bayesian inference using the maximum likelihood estimate, and an overall accuracy rate of 99.40% on the experimental data set.

AB - Finding a rapid gas identification algorithm with high accuracy and a closed-form solution that does not require any manual tuning of parameters is the major challenge to overcome in adopting electronic nose technology in daily life applications. Recently, bio-inspired rank-order-based classifiers have been proposed to meet this challenge by transforming multidimensional sensitivity vectors into temporally ordered spike sequences for target gases. The performance of these classifiers, however, is limited when the spike sequences corresponding to all the target gases do not contain sufficient discriminatory information to identify them. Moreover, their identification decision is delayed up to the computation of the sensitivity vectors at steady state, which incurs a long waiting time. In this paper, we adopt a Bayesian parametric method with a normal distribution model as an alternative approach that provides a closed-form solution with only second-order statistics, i.e., mean and covariance. However, for electronic nose applications, a reliable estimation of the covariance matrix is a major challenge with the commonly used maximum likelihood estimate in Bayesian inference because of the limited number of available measurements for each target gas. We exploit random matrix theory principles to reduce randomness in the sample covariance matrix for its reliable estimation. Moreover, transient features are used to accelerate the gas identification. In order to validate the effectiveness of this approach, data of eight gases, namely, C< ±0.5-ppm frequency stability3< ±0.5-ppm frequency stabilityH< ±0.5-ppm frequency stability8< ±0.5-ppm frequency stability, C< ±0.5-ppm frequency stability6< ±0.5-ppm frequency stabilityH< ±0.5-ppm frequency stability6< ±0.5-ppm frequency stability, CH< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stabilityO, CL< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stability, CO, CO< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stability, NO< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stability, and SO< ±0.5-ppm frequency stability2< ±0.5-ppm frequency stability, are acquired in the laboratory. We achieve a 7.73% performance improvement as compared with Bayesian inference using the maximum likelihood estimate, and an overall accuracy rate of 99.40% on the experimental data set.

KW - Bayesian inference

KW - electronic nose

KW - gas identification

KW - random matrix theory

UR - http://www.scopus.com/inward/record.url?scp=84962199107&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84962199107&partnerID=8YFLogxK

U2 - 10.1109/JSEN.2015.2507580

DO - 10.1109/JSEN.2015.2507580

M3 - Article

VL - 16

SP - 2036

EP - 2045

JO - IEEE Sensors Journal

JF - IEEE Sensors Journal

SN - 1530-437X

IS - 7

M1 - 7352294

ER -