Experimental verification of position sensing for a magnetic coil via Fano resonance

TL;DR

This paper shows how to locate a magnetic noise source using Fano resonance and machine learning, with high accuracy in distance and angle predictions.

Contribution

A novel method for position sensing of a magnetic coil using Fano resonance and supervised machine learning is experimentally verified.

Findings

The position of a transmitting coil can be predicted with a distance resolution of 0.048 m and an angular resolution of 8.8º.

Measured scattering parameters align well with analytical and numerical results.

The system works effectively for distances between 0.4 m and 2 m and angles within ± 60º.

Abstract

Predicting the position of a magnetic noise source in the kilohertz/megahertz range is of interest in academia and importance in industries. We consider a system consisting of one transmitting and four receiving coils around a frequency of 10 MHz. The transmitting coil is assumed to be a magnetic noise source that needs to be found in our study. We theoretically and experimentally show that the position of the transmitting coil can be predicted using Fano resonance, i.e., the four receiving coils are strongly coupled with each other, and weakly coupled to the transmitting coil via magnetic fields, by employing supervised machine learning. A coupled mode theory is built to elucidate such magnetic resonance coupling among coils. The characteristics of the system are investigated by using a method-of-moment based electromagnetic simulator, with sufficiently large volume of scattering parameters for different positions of the transmitting coil. Measured spectra of scattering parameters reasonably agree with analytical and numerical results. Experimental results reveal that the position of the transmitting coil is predicted in a range of 0.4 m to 2 m for the distance and ± 60º for the angle, with resolutions of 0.048 m and 8.8º, respectively.

Figures