[Paper Review] Resonant Very Low- and Ultra Low Frequency Digital Signal Reception Using a Portable Atomic Magnetometer

Stuart J. Ingleby, Iain C. Chalmers+3 more|arXiv (Cornell University)|Mar 6, 2020

Atomic and Subatomic Physics Research4 citations

TL;DR

This paper presents a portable, self-calibrating atomic magnetometer that enables resonant detection and decoding of digital signals at very low (VLF) and ultra-low (ULF) frequencies (200 Hz–200 kHz) using optical pumping of 133Cs vapor. It achieves bit error rates below 10% at received carrier powers down to -40 dBm, demonstrating sub-sea communication feasibility with projected ranges up to several kilometers.

ABSTRACT

Radio communication through attenuating media necessitates the use of very-low frequency (VLF) and ultra-low frequency (ULF) carrier bands, which are frequently used in underwater and under-ground communication applications. Quantum sensing techniques can be used to circumvent hard constraints on the size, weight and noise floor of classical signal transducers. In this low-frequency range, an optically pumped atomic sample can be used to detect carrier wave modulation resonant with ground-state Zeeman splitting of alkali atoms. Using a compact, self-calibrating system we demonstrate a resonant atomic transducer for digital data encoded using binary phase- and frequency-keying of resonant carrier waves in the 200 Hz -200 kHz range. We present field trial data showing sensor noise floor, decoded data and received bit error rate, and calculate the projected range of sub-sea communication using this device.

Motivation & Objective

• To develop a compact, field-deployable atomic magnetometer for resonant detection of VLF and ULF digital signals in conductive environments like seawater.
• To overcome limitations of classical inductive antennas, such as large size and high 1/f noise, by leveraging quantum sensing with atomic spin precession.
• To demonstrate practical, portable operation of a resonant atomic transducer outside laboratory settings with calibrated signal reception.
• To evaluate bit error rate (BER) performance across a wide frequency band (200 Hz–200 kHz) under real-world field conditions.
• To estimate the theoretical communication range for sub-sea applications based on measured signal-to-noise and attenuation characteristics.

Proposed method

• A compact, self-calibrating system uses a thermally stabilized vertical-cavity surface-emitting laser (VCSEL) to optically pump and probe a 133Cs vapor cell with 480 torr N2 buffer gas.
• Atomic spin precession is detected via magneto-optical rotation of the probe laser, with signal measured using a polarimeter and trans-impedance amplifier.
• Three-axis magnetic field coils compensate for Earth's field, enabling tunable Larmor frequency resonance across the 200 Hz–200 kHz range.
• Digital data is encoded using binary phase-shift keying (BPSK) and frequency-shift keying (FSK) on resonant carrier waves matched to the Zeeman splitting of 133Cs ground states.
• Signal amplitude is calibrated using the saturation of the 133Cs D1 transition, enabling SI-traceable power measurement.
• Homodyne detection and non-linear magneto-optical rotation are employed to maximize signal-to-noise ratio and suppress background noise.

Experimental results

Research questions

• RQ1 Can a portable, self-calibrating atomic magnetometer achieve resonant detection of digital signals in the VLF and ULF bands under real-world field conditions?
• RQ2 What is the achievable bit error rate (BER) for BPSK and FSK modulation across 200 Hz–200 kHz with received signal powers down to -100 dBm?
• RQ3 How does the sensor’s noise floor (5 pT·Hz⁻¹/²) compare to classical inductive sensors in practical, non-laboratory environments?
• RQ4 To what extent does off-resonant interference or non-Gaussian noise sources degrade BER performance across different carrier frequencies?
• RQ5 What is the projected communication range for sub-sea applications based on measured signal attenuation and sensor sensitivity?

Key findings

• The sensor achieved a noise floor of 5 pT·Hz⁻¹/² in non-laboratory field trials, demonstrating feasibility for portable operation.
• Bit error rates fell below 10% for received carrier powers not exceeding -40 dBm across the 200 Hz–200 kHz range.
• BER performance showed significant variation across frequencies without strong correlation to frequency change, suggesting dominant non-white or harmonic interference.
• A consistent excess BER was observed at 1 kbit/s symbol rate, attributed to spectral clipping by the sensor’s bandwidth limited by atomic relaxation rate Γ.
• Theoretical BER curves under additive white Gaussian noise (AWGN) were compared to measured data, showing deviation at lower signal powers.
• Projected communication ranges for sub-sea transmission were estimated based on seawater attenuation and measured sensitivity, indicating potential for multi-kilometer links with 10 W transmit power.