# Adaptive Algorithm for Fast 3D Characterization of Magnetic Sensors

## TL;DR

This paper introduces a fast algorithm for 3D characterization of magnetic sensors, improving accuracy and efficiency in modeling their spatial behavior.

## Contribution

A novel adaptive algorithm using contraction–expansion approach for faster and more accurate 3D magnetic sensor characterization.

## Key findings

- The proposed algorithm achieves faster convergence and smaller system distance between estimations and measurements.
- Runtimes average less than 1.5 seconds per direction for full impulse response estimation.
- The method enables feasible frequency and directivity characterization for magnetic sensors.

## Abstract

Magnetic sensors are highly relevant in clinical and industrial applications such as localization tasks and geological investigations. The spatial behavior of these sensors is of great interest for accurate forward modeling and the consequential possibilities for sophisticated applications, e.g., solutions to inverse problems. In this contribution, we present a novel characterization approach using adaptive system identification approaches. We utilize a gradient-based algorithm for estimating impulse and corresponding frequency responses for a directivity analysis in 1D, 2D, and 3D. For this, we built a triaxial Helmholtz coil setup to generate a 3D directive field. This is controlled by an algorithm that exploits similarities in sensor behavior with respect to small differences in excitation field angles. We found advantages for a controlled adaptation, with faster convergence and a smaller system distance between estimations and measurements with a proposed control based on the contraction–expansion approach (CEA). With runtimes averaging less than 1.5 s per direction for full impulse response estimation, this proof of concept shows the potential of the proposed algorithm for enabling a feasible frequency and directivity characterization method.