A Temperature-Compensated High-Resolution Microwave Sensor Using Artificial Neural Network

In this study, a loss-compensated microwave (MW) planar sensor is used to characterize fluids at ~1 GHz. The environmental temperature is shown to adversely impact the recorded resonance frequency of the MW sensor, leading to data mixing. This issue is resolved using a feedforward artificial neural...

Full description

Saved in:
Bibliographic Details
Published inIEEE microwave and wireless components letters Vol. 30; no. 9; pp. 1 - 4
Main Authors Kazemi, Nazli, Abdolrazzaghi, Mohammad, Musilek, Petr, Daneshmand, Mojgan
Format Journal Article
LanguageEnglish
Published IEEE 01.09.2020
Subjects
Artificial neural network (ANN)
machine learning
microwave (MW) sensor
Neurons
Resonant frequency
split-ring resonator (SRR)
temperature compensation
Temperature distribution
Temperature measurement
Temperature sensors
Online AccessGet full text

Cover

More Information
Summary:In this study, a loss-compensated microwave (MW) planar sensor is used to characterize fluids at ~1 GHz. The environmental temperature is shown to adversely impact the recorded resonance frequency of the MW sensor, leading to data mixing. This issue is resolved using a feedforward artificial neural network with two hidden layers. Various concentrations of methanol in water (0%-100% with 10% increments) are measured at temperatures ranging between 22 °C and 60 °C. This smart sensor system exhibits a strong ability to discriminate the correct data regardless of erroneous interfering factors up to 92%.
ISSN:1531-1309
1558-1764
DOI:10.1109/LMWC.2020.3012388