Adaptive Real-Time Magnetic Field Tracking beyond Prior Waveform Constraints

TL;DR

This paper presents an adaptive extended Kalman filter framework for real-time magnetic field estimation in quantum systems, overcoming noise constraints and improving tracking of weak, dynamic signals.

Contribution

The authors develop a novel adaptive estimation method that reduces model dependence and enhances real-time magnetic field tracking beyond traditional noise limits.

Findings

Successfully tracks weak magnetic signals beyond conventional sensitivity.

Validates the approach through numerical simulations and experiments.

Overcomes measurement noise constraints of standard filtering methods.

Abstract

The extraction of weak signals plays a crucial role in quantum precision measurement, where the estimation results are often limited by low signal-to-noise ratios. Here, we demonstrate a parameter-estimation framework based on the adaptive extended Kalman filter for dynamic magnetic-field estimation in quantum systems using spin-noise measurements -- a challenging regime characterized by weak signals. By modeling the magnetic field as an unknown parameter, the proposed approach alleviates model dependence in state estimation. Furthermore, by introducing an adaptive algorithm with real-time noise estimation, our method overcomes the measurement noise intensity constraints of conventional extended Kalman filtering and enhances its practical applicability. Numerical simulations covering three representative magnetic-field dynamics validate the capability of the proposed framework, while experimental results demonstrate successful tracking of a seismo-magnetic-like signal beyond the intrinsic sensitivity of conventional spin-noise spectroscopy.