Localization in indoor environments often requires expensive infrastructure that has to be maintained. One infrastructure-less method that can be used for distinguishing different rooms relies on measurements from ion-mobility spectrometry-based electronic noses (eNoses). These eNoses mimic the biological sense of smell and are widely used in artificial olfaction. In this talk we will show that it is possible to use eNoses also for localization. This means, that the characteristic combinations of airborne chemicals in different locations are used to distinguish between them. We will explain the working principle of ion-mobility spectrometry-based eNoses and how it can be used for positioning. The positioning accuracy of several machine learning algorithms under different measurement conditions will be presented and analyzed. Furthermore, potential questions and issues of eNose-based localization will be discussed, and potential solutions proposed.
Phillipp received his PhD degree from Tampere University of Technology, Finland, in 2016. Since 2021 he is a Senior Research Fellow at the Faculty of Information Technology and Communication, Tampere University, Finland, working on various projects that involve the use of inertial measurement units, GPS receivers, ion and differential mobility spectrometry devices. His field of interest is Bayesian Statistics and Filtering, and Machine Learning for the analytics of all kinds of sensor data.
Localization in indoor environments often requires expensive infrastructure that has to be maintained. One infrastructure-less method that can be used for distinguishing different rooms relies on measurements from ion-mobility spectrometry-based electronic noses (eNoses). These eNoses mimic the biological sense of smell and are widely used in artificial olfaction. In this talk we will show that it is possible to use eNoses also for localization. This means, that the characteristic combinations of airborne chemicals in different locations are used to distinguish between them. We will explain the working principle of ion-mobility spectrometry-based eNoses and how it can be used for positioning. The positioning accuracy of several machine learning algorithms under different measurement conditions will be presented and analyzed. Furthermore, potential questions and issues of eNose-based localization will be discussed, and potential solutions proposed.
Phillipp received his PhD degree from Tampere University of Technology, Finland, in 2016. Since 2021 he is a Senior Research Fellow at the Faculty of Information Technology and Communication, Tampere University, Finland, working on various projects that involve the use of inertial measurement units, GPS receivers, ion and differential mobility spectrometry devices. His field of interest is Bayesian Statistics and Filtering, and Machine Learning for the analytics of all kinds of sensor data.
