Journal
IEEE SENSORS LETTERS
Volume 4, Issue 10, Pages -Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LSENS.2020.3024606
Keywords
Chemical and biological sensors; electroantennogram (EAG); Hodgkin-Huxley (HH) model; real-time odor discrimination
Funding
- Mitsubishi Foundation
- JSPS KAKENHI [JP19K14943, JP19H02104, JP18H05467]
- National Bio-Resource Project (NBRP) of MEXT, Japan
Ask authors/readers for more resources
In this letter, a system that discriminates between two odorants in real-time with a single antenna of an insect was constructed. In previous studies, odorants were discriminated through arraying multiple types of sensors. However, the use of a sensor array enlarges the entire system and requires complex signal processing, which makes it difficult to mount the system on a quadcopter with limited payloads or an autonomous robot with low computational power. On the other hand, the antenna of an insect is composed of several olfactory receptors; thus, a single antenna responds to multiple odorants. Therefore, a system for discriminating between multiple odorants from the electroantennogram (EAG) signal of a single antenna of an insect is proposed. An antenna of an adult male silkmoth is employed; the silkmoth antenna exhibits an electric potential change when detecting an odor, but the difference in the EAG amplitude and recovery time depends on the type of odorant. Since EAG is a type of neural signal, it is difficult to perform odor discrimination by simply setting the threshold, because it drifts due to noise. Therefore, the Hodgkin-Huxley model (HH model) was applied to the raw EAG signal as a dynamic filter, and the spike firing rate of the output value was calculated using the HH model. Then, the odorants were discriminated against based on the magnitude of the spike firing rate. A series of signal processing was implemented in a microcontroller, and the experiments indicated that all antennae demonstrated a discrimination performance of 90% or more.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available