Joint Distribution Adaptation for Drift Correction in Electronic Nose Type Sensor Arrays

Jersson X. Leon-Medina,Diego Alexander Tibaduiza Burgos,Wilman Alonso Pineda-Munoz

doi:10.1109/access.2020.3010711

Abstract

This research deal with the drift compensation problem in sensor arrays named electronic noses. The drift problem occurs in this kind of sensor when they are exposed to an analyte for long periods, which may cause that the response of the sensor varies with time. Some approaches in the literature have tackled the drift compensation problem from the point of view of signal processing algorithms to obtain high rates of accuracy independently of time. In this work, the drift problem is solved using transfer learning with the joint distribution adaptation (JDA) method, which adapts both marginal and conditional distributions between domains, and requires no labeled data in the target domain to perform a classification task with a machine learning algorithm. The developed methodology for drift compensation is validated by measuring accuracy in the classification process. Validation considers a data set that measured six volatile organic compounds during a period of three years under strongly controlled operating conditions using a series of 16 metal oxide gas (MOX) sensors. JDA and Kernel JDA are used with three different types of kernels to determine the best behavior in terms of accuracy to correct the drift in electronic noses. As a result, it can be concluded that the approach using JDA outperforms standard learners like K-Nearest Neighbor (KNN) method.

Highlights

Electronic noses are an array of sensors that can be defined as a collection of cross-sensitivity sensors connected to an electronic data acquisition unit, and a pattern recognition system
The number of subspace bases k refers to the size of the embedding matrix Z obtained after performing the Principal Component Analysis (PCA) to construct a new shared subspace. k can be chosen so that the low-dimensional representation is accurate for data reconstruction
The Drift problem presented in electronic noses was solved using a methodology based on transfer learning with the joint distribution adaptation (JDA) method

Summary

INTRODUCTION

Electronic noses are an array of sensors that can be defined as a collection of cross-sensitivity sensors connected to an electronic data acquisition unit, and a pattern recognition system. Several studies have tried to solve the drift problem in electronic noses with machine learning techniques This is the case of Liu et al in 2013 [9], who developed a drift compensation method based on semi-supervised domain adaption. For all the above reasons, in this work a drift correction method that exploits the benefits of the Joint Distribution Adaptation (JDA) was developed This algorithm of transfer learning adapts both marginal and conditional distributions between domains, no labeled data are required in the target domain and the reliability of electronic nose sensors arrays can be increased in the long term.

THEORETICAL BACKGROUND

DATASET FOR VALIDATION

EXPERIMENTAL SETTINGS

PERFORMANCE VERIFICATION

CONCLUSIONS