Compressed sensing enabled high-bandwidth and large dynamic range magnetic sensing

TL;DR

This paper introduces a compressed sensing approach to magnetic sensing with NV centers, significantly reducing data requirements and improving measurement accuracy in low SNR conditions, with broad applicability in quantum sensing.

Contribution

The study presents a novel application of compressed sensing in ESR-based magnetic sensing, demonstrating improved accuracy and efficiency over traditional raster scanning methods.

Findings

3x improvement in measurement accuracy in low SNR data

Achieves same accuracy with only 15% of data points

Applicable to ESR measurements beyond NV centers

Abstract

Electron Spin Resonance (ESR) is a widely common method in the field of quantum sensing. Specifically with the Nitrogen-Vacancy (NV) center in diamond, used for sensing magnetic and electric fields, strain and temperature. However, ESR measurements are limited in temporal resolution, primarily due to the large number of data points required especially in high dynamic range regimes and the need for extensive averaging caused by low signal-to-noise ratio (SNR). This study introduces a novel application of compressed sensing (CS) for magnetic sensing using NV centers. By comparing CS with conventional raster scanning, we demonstrate the potential of CS to enhance sensing applications. Experimental results, supported by simulations, show an improvement of factor 3 in measurement accuracy in low SNR data, which also translates to achieving the same accuracy with only 15% of the data points. Moreover, the proposed approach is not confined to NV centers but can be extended to ESR measurements in other systems, broadening its applicability in quantum sensing.