Sub-femtosecond stabilization of multicore fiber for high-fidelity quantum networking at 100% duty cycle

By leveraging the high noise correlation between cores in a 40-km multicore fiber, researchers achieved sub-femtosecond stabilization that enables 100% duty cycle quantum networking with negligible crosstalk-induced spurious photons.

The Gist

Imagine you are trying to send a incredibly fragile, secret message (a "quantum" signal) through a long, bumpy tunnel. At the same time, you need to send a loud, bright flashlight beam (a "classical" signal) through the same tunnel to measure how much the tunnel is shaking, so you can cancel out the shaking and keep your secret message steady.

The problem is that the bright flashlight is so loud that it drowns out the secret message, or its light leaks into the secret message's path, creating "noise" (like static on a radio). Usually, to stop this, you have to take turns sending the messages (turning the flashlight off while sending the secret message), which slows everything down, or you use heavy filters that lose some of the secret message along the way.

The Solution: A Multi-Lane Highway
This paper introduces a special type of cable called a Multicore Fiber (MCF). Think of this not as a single tunnel, but as a 7-lane highway all bundled inside one giant tube.

The researchers used two specific lanes (cores) of this highway:

1. Lane A (The Quantum Lane): Carries the fragile secret message.
2. Lane B (The Stabilization Lane): Carries the bright flashlight beam used to measure the shaking.

Why This Works: The "Twin" Effect
Even though the lanes are separate, they are packed so tightly together inside the same tube that they react to the environment almost exactly the same way. If the ground shakes, Lane A and Lane B shake in perfect sync. They are like identical twins walking side-by-side; if one stumbles, the other stumbles at the exact same moment.

Because they shake together, the researchers can listen to the "stumbling" in Lane B (the bright beam) and instantly tell Lane A (the secret message) how to adjust to stay perfectly still. This allows them to keep the secret message perfectly synchronized without ever having to turn off the flashlight. They can send both signals 100% of the time, with no pauses.

The Results: Silence in the Noise
The team tested this over a 40-kilometer (about 25-mile) spool of this special fiber. Here is what they achieved:

• Perfect Timing: They stabilized the timing of the secret message to within 100 attoseconds. To put that in perspective, an attosecond is to a second what a second is to the age of the universe. It is an almost unimaginable level of precision.
• No "Leakage": Usually, the bright light from Lane B would leak into Lane A, creating "ghost" photons (noise) that ruin the secret message. However, because the lanes are so well-separated and the researchers used a specific color of light for the flashlight that is slightly different from the secret message, the leakage was incredibly low.
• The "Ghost" Count: They calculated that the number of unwanted "ghost" photons leaking into the secret lane was less than 0.01 per second. That is essentially zero. It is so quiet that the "noise" from the fiber is lower than the natural background noise of the detector itself.

The Big Picture
The paper demonstrates that by using this "7-lane highway" approach, we can finally send quantum information and classical data together over long distances without them interfering with each other. This allows for a quantum network that is always "on" (100% duty cycle), incredibly stable, and free from the noise that usually plagues these systems. It's a major step toward building a future internet where quantum information can travel reliably and securely.

Technical Summary

1. Problem Statement

The co-existence of quantum and classical light within the same optical fiber is a critical bottleneck for real-world quantum networks.

• The Challenge: Quantum networking protocols require sub-femtosecond (fs) synchronization to cancel fiber-induced phase noise. This typically requires a strong classical "stabilization" beam to co-propagate with the weak quantum signal.
• Limitations of Current Solutions:
  • Single-Mode Fiber (SMF) with WDM/TDM: Using different wavelengths (Wavelength Division Multiplexing) or time slots (Time Division Multiplexing) in a single core introduces significant issues. WDM suffers from Spontaneous Raman Scattering (SRS), where photons from the bright classical beam scatter into the quantum channel, creating noise. TDM reduces the duty cycle of the quantum channel, lowering throughput.
  • Fiber Pairs: Using separate fibers for quantum and classical signals fails because environmental noise (temperature, vibration) is not sufficiently correlated between two distinct fibers to allow for effective noise cancellation.

2. Methodology

The authors propose and experimentally validate a solution using Multicore Fiber (MCF) to achieve high-fidelity co-existence without sacrificing duty cycle or increasing loss.

• Experimental Setup:
  • Fiber: A 40 km spool of weakly coupled 7-core MCF (cladding diameter 160 μm).
  • Channel Separation:
    • Quantum Channel: Core-2, operating at 1542 nm (simulating a quantum signal).
    • Stabilization Channel: Core-7, operating at 1550 nm (classical synchronization light).
  • Noise Correlation: The cores are confined within the same cladding, ensuring that environmental noise affects both cores almost identically. This allows the phase noise measured in Core-7 to be used to actively stabilize Core-2.
  • Active Stabilization:
    • An Acousto-Optic Modulator (AOM) system creates a heterodyne beat note for Core-7.
    • A feedback loop locks the phase of Core-7 to a reference.
    • The same correction signal is applied to Core-2 (via a shared AOM driver) to cancel fiber-induced phase noise.
  • Low-Power Strategy: To minimize Raman scattering, the authors optimized the system to use the minimum possible optical power for stabilization while maintaining a "cycle-slip-free" lock.

3. Key Contributions

• Sub-femtosecond Jitter over 40 km: Demonstrated integrated timing jitter of 100 attoseconds (after removing source laser noise) over a 40 km link, a level previously unachieved in MCF for quantum networking.
• 100% Duty Cycle: Unlike TDM schemes, this approach allows the quantum channel to operate continuously.
• Quantification of Raman Noise: Provided the first quantitative measurement of spontaneous Raman scattering photons coupled between MCF cores, proving that the noise floor is negligible.
• Ultra-Low Power Stabilization: Demonstrated that robust phase locking can be achieved with only 400 pW of launched stabilization power (resulting in ~300 fW at the detector), drastically reducing the source of Raman noise.

4. Key Results

A. Phase Noise and Stability

• Integrated Jitter:
  • 400 attoseconds (1 MHz to 0.5 Hz) for the stabilized quantum channel (Core-2).
  • 100 attoseconds when the integration bandwidth is limited to 1 kHz–0.5 Hz (removing source laser noise and servo bump contributions).
• Long-Term Stability:
  • Over 6 hours, the system maintained cycle-slip-free stabilization.
  • Peak-to-peak timing errors were reduced from 10 ps (free-running) to **25 fs** (stabilized).
  • Time Deviation (TDEV) remained below 1 fs for averaging times up to several hundred seconds.
• Noise Suppression: Achieved >20 dB noise suppression for offset frequencies below 10 Hz. The residual noise is limited by the finite correlation between cores and the wavelength difference (8 nm detuning).

B. Raman Scattering and Spurious Photons

• Crosstalk: Measured core-to-core crosstalk at -40 dB.
• Spurious Photon Rate:
  • Using the low-power stabilization (400 pW) and the 40 dB isolation, the estimated rate of Raman-scattered photons entering the quantum channel is <0.01 photons/s (in a 100 GHz bandwidth).
  • Experimental verification using a Superconducting Nanowire Single Photon Detector (SNSPD) confirmed this rate, which is 4 orders of magnitude lower than what is achieved in single-core SMF with WDM.
  • This rate is comparable to the dark count rate of state-of-the-art detectors, meaning the stabilization light does not significantly degrade the signal-to-noise ratio.

5. Significance

This work establishes Multicore Fiber as a superior infrastructure for future quantum networks.

• High Fidelity: By reducing spurious photon counts to near-zero levels, the method preserves the fidelity of quantum states (e.g., for Quantum Key Distribution).
• Scalability: The ability to run 100% duty cycle quantum channels alongside high-power classical synchronization signals enables high-throughput, long-distance quantum networks.
• Practicality: The demonstration of sub-femtosecond stability over 40 km with low-power requirements suggests that this technology is viable for real-world deployment, overcoming the "capacity crunch" and noise limitations of current single-fiber solutions.

In summary, the paper proves that MCF can simultaneously solve the problems of timing synchronization (via noise correlation) and optical crosstalk (via spatial separation), enabling ultra-stable, high-fidelity quantum networking.