Proof-of-principle long-distance Sagnac twin-field quantum key distribution network

arXiv:2609.04447 · quant-ph, physics.app-ph, physics.optics · Submitted 2026-09-03 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: I'm Kai, and with me are Mira and Lev, guest researcher.

Mira: Today's paper: "Proof-of-principle long-distance Sagnac twin-field quantum key distribution network".

Kai: Twin-field (TF) quantum key distribution (QKD) offers a promising approach to long-distance QKD networks due to its superior performance over large channel losses.

Mira: First, who's behind it and why it matters.

Title and authors: Kai: So we’re looking at the title "Proof-of-principle long-distance Sagnac twin-field quantum key distribution network," which immediately tells us this is an experimental demonstration focusing on establishing feasibility over distance using TF technology.

Mira: The authors, Mandil, Shih, Verma, Qian, and Lo, represent a strong mix of hardware expertise and theoretical grounding necessary for such a complex setup. I’m curious how their specific backgrounds informed the choice of the Sagnac topology over other ring configurations.

Lev: From an error correction standpoint, having researchers from different institutions like Toronto and National Tsing Hua University suggests they’re looking at this problem from several distinct angles, which is usually good for finding hidden assumptions.

Kai: The implication here is that they’re showing how TFQKD can be made practical for long-haul links without relying on the massive infrastructure needed for active phase stabilization.

Mira: It implies that the fundamental physics of twin-field states can be exploited effectively even when dealing with high channel loss scenarios, provided you have a robust method to handle the resulting asymmetry in transmission.

Lev: It suggests that error correction techniques, like those I work on, might need to adapt their assumptions about channel characteristics when scaling up from laboratory demonstrations to actual hardware.

Kai: So, essentially they’re proving that the architecture itself is sound enough for long-distance communication using these specific components without needing external active stabilization.

Mira: Exactly; it's a proof of concept that the lossy channel problem can be managed by clever network geometry and protocol design rather than just brute-force amplification.

Lev: It sets a benchmark for what kind of hardware robustness we need to consider if we want to move these long-distance TFQKD links into operational use.

The paper's summary: Kai: The summary of "Proof-of-principle long-distance Sagnac twin-field quantum key distribution network" boils down to them demonstrating a three-user pair Sagnac TFQKD network spanning one hundred twenty-seven kilometers using single-photon avalanche detectors without any active phase stabilization or postcompensation.

Mira: That's the main takeaway, Kai; they achieved this long distance and multi-user setup while explicitly avoiding the typical hardware hurdles associated with maintaining interferometric stability over such lengths.

Lev: What’s interesting is that they managed to integrate the finite-size security analysis into a protocol designed for asymmetric channels between Charlie and each user, which is a significant extension of prior TFQKD work.

Kai: They use a ring topology where Alice, Bob, and Danny are connected by long fibers of unequal lengths, utilizing pulses generated by an on-off burst pattern through an intensity modulator and variable optical attenuator.

Mira: That burst patterning seems crucial because it directly addresses the challenge of high signal-to-noise ratio when Rayleigh backscattering noise is present in those long channels.

Lev: I’m thinking about how they handle the interference at Charlie’s station, specifically using a fifty:fifty beam splitter followed by two detectors to confirm successful measurement events.

Kai: The protocol involves users selecting bases—X for signal with probability PX, and Z for decoy states with probability PZ equals one minus PX—and preparing specific quantum states accordingly.

Mira: Their description of the five-step process, especially how they combine X basis key generation with Z basis decoy estimation for phase error rate measurement, is where the theoretical meat of their security analysis lies.

Lev: That combination allows them to get both a raw key and a continuous assessment of link quality simultaneously, which is a powerful diagnostic tool.

The paper's improvements: Kai: The paper suggests several improvements centered around the active polarization control scheme they implemented, which uses electronic polarization controllers driven by feedback from detectors like DA, DB1, DB2, and DD to monitor intensity.

Mira: That feedback loop is essentially an attempt to dynamically compensate for fiber birefringence fluctuations in real-time; it’s moving beyond static alignment towards adaptive control.

Lev: If that monitoring system can tune the EPC settings based on those photon counts every five hundred milliseconds, then we have a dynamic method to maintain the necessary interference visibility.

Kai: They also highlight a theoretical result showing that if certain conditions on parameters A2 through A7 are met, then all of them are satisfied, which simplifies the control system because only one user needs to monitor both directions.

Mira: That theoretical simplification is neat; it suggests that the complexity of controlling polarization might be significantly reduced if you can satisfy those specific mathematical constraints.

Lev: From a hardware perspective, simplifying the monitoring requirements means fewer components to build and less potential for noise injection into the feedback loop, which is always a win in experimental physics.

Conclusion: Kai: So to wrap up on "Proof-of-principle long-distance Sagnac twin-field quantum key distribution network," the paper successfully demonstrated a working, lossy TFQKD network over one hundred twenty-seven km without requiring active phase stabilization or postcompensation.

Mira: The implication is that this work provides a tangible path toward scaling up QKD networks by showing that clever topological design and protocol adaptation can overcome significant channel loss challenges.

Lev: For the error correction community, it serves as a crucial reminder that hardware robustness isn't just about achieving a high raw rate; it’s about managing the continuous noise floor to maintain security over long spans.

Kai: It really shows that with careful engineering, we can push these quantum communication systems further into regimes that were previously unreachable due to stabilization requirements.

Mira: Indeed, the ability to achieve positive key rates in the finite data-size regime is a strong indicator that this approach has practical implications for deployment when considering network scaling and multi-user operation.

Lev: I just think we need to keep pushing on developing the error correction methods that can handle these dynamically controlled channels if we want this technology to become a reality.

Centre for Quantum Information and Quantum Control, Department of Physics, University of Toronto · Centre for Quantum Information and Quantum Control, Department of Electrical and Computer Engineering, University of Toronto · Department of Physics and Center for Quantum Science and Technology, National Tsing Hua University · Department of Physics, National University of Singapore · Centre for Quantum Technologies, National University of Singapore · Quantum Bridge Technologies, Inc.

quant-ph, physics.app-ph, physics.optics

Submitted: 2026-09-03

Updated: 2026-10-01

Comments: 9 pages, 5 figures

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 72/100

The gist: Twin-field (TF) quantum key distribution (QKD) offers a promising approach to long-distance QKD networks due to its superior performance over large channel losses.

Key concepts

Twin-field (TF) QKD
A quantum key distribution method that uses two independent optical fields to transmit information. This approach is highly effective for long distances because it mitigates the effects of large channel losses, making it more robust than traditional QKD systems over long fibers.
Sagnac Network
A ring topology network where light travels in a closed loop. In this experiment, three users (Alice, Bob, Danny) are connected in a circle. This configuration allows for the counterpropagating pulses to interact constructively at the central relay (Charlie), which is crucial for the network's operation.
Burst-patterning
A technique used to structure the light pulses so that they achieve a high signal-to-noise ratio, even when there is significant Rayleigh backscattering noise in the fiber. This method ensures that each active user modulates only one path at a time without causing pulse overlap.
Polarization Stabilization Scheme
A system designed to keep the polarization of light stable over long distances. It uses active feedback loops and electronic polarization controllers (EPCs) to monitor light intensity and automatically adjust the fiber segments' alignment, ensuring high interference visibility for accurate key generation.

Terminology

Summary

Twin-field (TF) quantum key distribution (QKD) offers a promising approach to long-distance QKD networks due to its superior performance over large channel losses. The gist is: In this work, we experimentally demonstrate a proof-of-principle three-user-pair Sagnac TFQKD network spanning 127-km using single-photon avalanche detectors without any active phase stabilization or postcompensation. This research is significant because it presents a highly practical and cost-effective approach to long-distance QKD networks, overcoming challenges associated with lossy channels and multi-user scaling.

Network Architecture and Configuration

The experiment demonstrates a three-user-pair Sagnac TFQKD network spanning 127-km. The network utilizes a ring topology where three users—Alice, Bob, and Danny—are separated by long fibers of unequal lengths. A key feature is the use of a continuous-wave distributed feedback laser (1545.3 nm) which passes through an intensity modulator (IM) and a variable optical attenuator (VOA). The light is launched into the fiber ring via a symmetric beam splitter.

The network operates with burst-patterning necessary to achieve high signal-to-noise ratio (SNR) in the presence of Rayleigh backscattering noise. In a given key generation session, each active user modulates only one path, specifically the one that has already traversed the other active user. For instance, in an Alice-Bob pair, Alice will modulate the counterclockwise traveling pulses while Bob modulates the clockwise traveling pulses. The fiber length at each user station can be adjusted such that counterpropagating pulses never overlap at any modulator.

Protocol and Key Generation

The protocol adopted is a five-step process based on finite-size security analysis, extending the TFQKD protocol to asymmetric channels between Charlie and each user. The steps include:

  1. Two users independently select a basis, choosing the X (signal) basis with probability PX and the Z (decoy) basis with probability PZ = 1 − PX. If the X basis is chosen, the user prepares a WCP with a preselected global phase and intensity s according to probability Ps = PX. The user selects between a bit value of 0 or 1 at random, to add a 0 or π phase, respectively, to their WCP. If the Z basis is chosen, the user prepares a phaserandomized WCP with intensity chosen from a set of decoy settings and probabilities Pµ, Pν.

  2. The two users send their prepared WCPs to an untrusted central relay, Charlie.

  3. At Charlie’s station, the incoming WCPs interfere with each other at a 50:50 beam splitter followed by two single-photon detectors, D0 and D1. A measurement event is considered successful if one and only one detector clicks.

  4. Charlie announces the results of the successful measurement events (i.e., which detector clicked) and the two users declare which basis they selected.

  5. Based on this information, the two users distill the secret key, using instances where both chose X for raw key generation and Z for decoy-state method estimation of phase error rate.

Polarization Stabilization Scheme

To maximize interference visibility and ensure accurate encoding, active polarization control is essential to maintain long-term interferometric stability. The scheme involves:

  1. Each user employs a PM fitted with a polarizer and an IM with a polarization-maintaining fiber input to guarantee proper alignment inside the modulators.

  2. Fiber segments between users are polarization-aligned using electronic polarization controllers (EPCs).

  3. Intensity of light passing through the user station is monitored using a fiber tap and SPAD, and this signal drives an EPC to adjust its setting to increase intensity. For example, Alice monitors clockwise traveling light with detector DA to drive EPC1 in the segment between Charlie and Alice.

The theory shows that If A2 and A5 and one of A3 and A7 and one of A4 and A6 are satisfied, then all of A1–A7 are satisfied. This allows for a system where only one user in the network needs to monitor both directions. The automated feedback control system uses a modified gradient-ascent optimization algorithm to tune the EPCs based on photon counts at detectors DA, DB1, DB2, and DD every 500-ms.

Experimental Results and Performance

The experiment achieved a stable Sagnac interference visibility of 93 ± 1% over one hour. The key rate achieved over an asymmetric communication channel with 102-km fiber and 45-dB overall loss is 1.398 × 10−5 bits per pulse. This result represents the first TFQKD network without active phase stabilization or postcompensation achieved over long fibers.

Improvements for AI systems

As a fastidious and diligent researcher, I have analyzed this proof-of-principle work on a long-distance Twin-Field Quantum Key Distribution (TFQKD) network. While the paper demonstrates a novel Sagnac configuration without active phase stabilization or postcompensation over 127 km, its primary contribution lies in establishing feasibility for low-cost, scalable, multi-user quantum networks.

Improving AI systems using this scientific paper involves leveraging the principles of robust quantum communication and network management to enhance specific AI capabilities.

Here are the specific improvements and what the improved AI system can do:

)1. Robust Quantum Channel Characterization (Improved Perception/Sensing Module)

The paper details a sophisticated active polarization control scheme driven by real-time intensity monitoring (using SPAD counts DA, DB1, DB2, DD). This provides high-fidelity feedback on fiber birefringence fluctuations and polarization drift over long distances.

The improved AI system can perform:

  • Detect and predict fiber channel impairments (birefringence changes) in real-time with sub-second latency.

  • Dynamically adjust the control parameters (e.g., EPC settings) to maintain maximum interference visibility (target 93±1%) even as environmental noise fluctuates.

)2. Asymmetric Channel Key Rate Optimization (Improved Decision/Optimization Module)

The protocol handles asymmetric communication channels between Charlie and each user, where signal intensities vary, and the key rate depends on complex probability distributions of decoy states and basis choices.

The improved AI system can perform:

  • Optimize the sending probabilities for signal versus decoy states to maximize the secure key rate under varying channel loss conditions (e.g., 45-dB loss in the example).

  • Predict optimal intensity settings for WCP preparation based on real-time channel attenuation measurements, ensuring the most efficient use of limited quantum resources.

)3. Network Topology and User Management (Improved Planning/Resource Allocation Module)

The architecture supports a three-user network where any two users can exchange keys while the third remains inactive, and users can be added by inserting stations along the ring.

The improved AI system can perform:

  • Automated network topology reconfiguration in response to node failures or desired key generation requirements.

  • Dynamic resource allocation for multi-user tasks (e.g., prioritizing Alice+Bob key generation over Alice+Danny) based on predicted channel performance and required security levels.

)4. Noise Mitigation and Error Rate Estimation (Improved Diagnostics/Filtering Module)

The protocol uses the Z basis measurements to estimate the phase error rate, while the X basis measures QBER, allowing for comprehensive security analysis in both infinite-data and finite-data regimes.

The improved AI system can perform:

  • Distinguish between quantum bit errors (QBER) and classical noise sources (Rayleigh backscattering, dark counts) by analyzing the gains across different measurement bases.

  • Accurately estimate the phase error rate in real-time to provide a continuous security assessment of the link quality, preventing key distillation if security thresholds are breached.

This synthesis transforms a static QKD demonstration into an intelligent, self-optimizing quantum network management system capable of maintaining high performance over long, lossy fiber links without requiring expensive active phase stabilization hardware.

Abstract

Twin-field (TF) quantum key distribution (QKD) offers a promising approach to long-distance QKD networks due to its superior performance over large channel losses. Due to specialized hardware requirements, nearly all long-distance TFQKD demonstrations have only two users exchanging keys, rather than a network with three or more users. In this work, we experimentally demonstrate a proof-of-principle three-user-pair Sagnac TFQKD network spanning 127-km using single-photon avalanche detectors without any active phase stabilization or postcompensation. We implement efficient procedures for maintaining polarization stability and circumventing Rayleigh backscattering noise to achieve a stable Sagnac interference visibility of 93 plus or minus1 % over one hour. A secure key rate of 1.398 times10-5 bits per pulse is achieved over an asymmetric communication channel with 102-km fiber and 45-dB overall loss. To our knowledge, this is the first TFQKD network without active phase stabilization or postcompensation achieved over long fibers. Our results represent a highly practical and cost-effective approach to long-distance QKD networks.

Related papers