Dipole Localization Using An Integrated Radio Frequency Atomic Magnetometer

TL;DR

This work introduces a method to localize a nearby RF source modeled as an oriented magnetic dipole using two vector measurements from a single integrated RF atomic magnetometer, enabling true three-dimensional localization in unshielded environments. By leveraging the dipole field relation and two directional measurements with orthogonal pump configurations, the dipole direction vectors are obtained and their intersection yields the source position, independent of absolute dipole strength. The authors validate the approach with a controlled dipole source at 423 kHz and develop model-deviation metrics MD1–MD3 to verify the dipole hypothesis, demonstrating good agreement between measured and predicted fields and robust 3D localization. The integrated magnetometer’s compact head, tunable resonance, and phase-sensitive detection enable field-ready dipole tracking for known-orientation sources in applications such as NMR, MIT, or RF tagging in unshielded environments.

Abstract

Optically-pumped atomic magnetometers have previously been used in arrays to reject interference from far away sources and enable the sensitive detection of local sources of radio frequency (RF) signals, useful, for instance, in the detection of low field NMR signals in an unshielded environment. We now demonstrate a complementary scheme in which four magnetometer measurements are used to locate in three dimensions a nearby radio frequency source. The methodology relies on the measurement of a radio frequency vector at two different positions and modeling the source as a magnetic dipole of known orientation. In contrast to coil detection, magnetometers have the advantage of measuring signals in a 2D plane, and do not inductively couple to their environment or each other, making them a strong candidate for localization of hidden RF sources. For this demonstration, we use only a single RF magnetometer to make four measurements of a synchronous and oriented dipole source, but it is to be expected that this could be replaced by four magnetometers working simultaneously. In addition, this work is greatly aided by the introduction of a fully integrated magnetometer, in which all optics, including lasers, are safely enclosed into a compact head with flexible wired connections. The portable, as well as safe, nature of the sensor make it quite valuable for in the field work.

Figures (10)

• Figure 1: Magnetic field lines from an idealized magnetic dipole.The angles $\theta$ and $\phi$ are defined as in the inset, while the unit direction vector $\hat{n}$ is from the magnetometer to the dipole.
• Figure 2: Vectorial schematic of the dipole position with respect to $\mathrm S_{1}$ and $\mathrm S_{2}$ representing the positions of the alkali metal cell of each sensor and the origin is taken on the x-axis at a position equidistant from $\mathrm S_{1}$ and $\mathrm S_{2}$. The x-axis is uniquely defined as the direction between the two sensors.
• Figure 3: a) This schematic shows a top view of the sensor, with an atomic cell (yellow), crossed pump (blue) and probe (red) laser beams, and field coils (brown) creating the bias field (green). b) All these components are enclosed in a cylindrical black tube. Also shown is a bench-top current controller, and the differential amplifier for use with the balanced polarimeter.
• Figure 4: Noise spectra measured at the resonance frequency of 423 kHz (NQR of ammonium nitrate) with an acquisition time of 2.048 ms. Signals obtained with an active RF transmitter driving the magnetometer with 0.2 nT resonant excitation are shown in red (inset). Spectrometer noise (black), with additional electronic noise (blue), and further added probe beam noise (green) exhibit similar behavior both (b) with and (a) without shielding. Magnetic noise or sensitivity (purple), observed with optically pumped atoms, is markedly higher without shielding due to environmental laboratory noise.
• Figure 5: a) Photo of the sensor head, alignment platform, dipole and calibration coils. b) A small coil with AC current running through it at 423 kHz is moved along a ruler above the plane containing the magnetometer positions. Measurement was taken in the positions shown, as well as with the sensor rotated by 90 degrees. c) Top view of measurement points and sensor positions during dipole movements along (i) only the x-axis, (ii) only the y-axis and (ii) the diagonal x/y-axis.