Scholarlink

This paper presents an all-optical setup utilizing NV-rich diamonds for magnetic field measurements in the frequency domain. The study focuses on resolving the low-field ambiguity that occurs in the magnetic field range up to 8 mT. Two approaches are examined to address this issue. The first employs non-linear least squares fitting, offering a clear solution without machine learning. The second approach uses a fully connected neural network, which achieves higher accuracy in predicting the magnetic field. In order to optimize the network for a possible implementation in microcontrollers, the size of the neural network and the used data are considered more closely. A comparative analysis will outline the strengths and limitations of each method.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Computational Approaches for Resolving the Low-Field Ambiguity in All-Optical Magnetic Field Sensing With NV Centers

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ann-Sophie Bülter,
  • Ludwig Horsthemke,
  • José Luis Ávila-Jiménez,
  • Frederik Hoffmann,
  • Francisco Javier Rodriguez-Lozano,
  • Sarah Kirschke,
  • Tilman Sanders,
  • Markus Gregor,
  • Peter Glösekötter

摘要

This paper presents an all-optical setup utilizing NV-rich diamonds for magnetic field measurements in the frequency domain. The study focuses on resolving the low-field ambiguity that occurs in the magnetic field range up to 8 mT. Two approaches are examined to address this issue. The first employs non-linear least squares fitting, offering a clear solution without machine learning. The second approach uses a fully connected neural network, which achieves higher accuracy in predicting the magnetic field. In order to optimize the network for a possible implementation in microcontrollers, the size of the neural network and the used data are considered more closely. A comparative analysis will outline the strengths and limitations of each method.