Dual-Pilot-Assisted Frequency and Phase Recovery in Experimental CV-QKD
summary
The gist
Continuous-variable quantum key distribution (CV-QKD) promises high rates and seamless integration with classical beams within a single optical fiber This work presents an end-to-end experimental…
In short
This work experimentally implemented continuous-variable quantum key distribution (CV-QKD) over a 12 km optical fiber link. The system used dual-pilot tones for robust frequency and phase recovery to accurately estimate channel parameters. The experiment successfully generated secret key rates of approximately 5.11 Mbit/s asymptotically and 4.67 Mbit/s in the finite-size regime, providing a practical baseline for metropolitan CV-QKD deployment.
Key concepts
- Continuous-Variable Quantum Key Distribution (CV-QKD)
- CV-QKD is a quantum communication protocol that uses continuous variables, such as the amplitude and phase of light, to establish a secure key between two parties. Unlike discrete systems, it transmits information using the properties of light waves rather than individual photons.
- Dual-Pilot-Assisted Recovery
- This technique involves sending specific reference signals called pilot tones alongside the quantum data. The system uses these known tones to precisely measure and correct for frequency offsets and phase mismatches that occur in the optical channel, ensuring accurate measurement of the quantum state.
- Information Reconciliation
- Since imperfect measurements introduce errors, information reconciliation is a classical post-processing step where Alice and Bob compare their data. They use sophisticated error-correcting codes to identify and fix discrepancies caused by noise, allowing them to agree on the final secret key.
Terminology used across episodes
This episode discusses
- Dual-Pilot-Assisted Frequency and Phase Recovery in Experimental CV-QKD · Paper Radio
- Cyber Security Awareness Campaigns: Why do they fail to change behaviour?
- Quantum cryptography: Public key distribution and coin tossing
- Continuous variable quantum cryptography using coherent states
- Coexistence of continuous-variable quantum key distribution and classical data over 120-km fiber
- Passive polarization and phase stabilization scheme for Twin-Field QKD
The paper
Dual-Pilot-Assisted Frequency and Phase Recovery in Experimental CV-QKD · Read on arXiv
Christiano M. S. Nascimento, Artur A. Matoso, Gustavo C. Amaral, Guilherme P. Temporão
QuIIN - Quantum Industrial Innovation EMBRAPII CIMATEC Competence Center in Quantum Technologies · NITeQ Department of Electrical Engineering Pontifical Catholic University of Rio de Janeiro · Quantum Technology Department TNO—The Netherlands Organization for Applied Scientific Research
Continuous-variable quantum key distribution (CV-QKD) offers high secret-key rates and compatibility with optical communication infrastructure. Locally generated local oscillator (LLO) architectures reduce security vulnerabilities associated with transmitting a strong optical reference, but independent lasers introduce frequency offsets and phase fluctuations that require accurate compensation. Here, we experimentally demonstrate a Gaussian-modulated LLO CV-QKD system over 12 km of optical fiber using two frequency-multiplexed pilot tones and RF-heterodyne detection with independent lasers. The proposed digital signal processing (DSP) architecture integrates CAZAC-based frame synchronization, global and subframe-wise carrier-frequency-offset estimation, and coherent dual-pilot phase recovery. We analyze 120 experimental frames, each containing approximately 10 6 coherent states, and directly compare single-pilot and dual-pilot processing using identical acquisitions. The dual-pilot DSP chain achieves an average input-referred excess noise of 15.56 plus or minus 7.61 mSNU, representing reductions of 28.1% and 17.4% relative to pilots 1 and 2, respectively. Under a trusted-receiver security model, the system achieves average modeled secret-key rates of 14.98 plus or minus 4.18 Mbit/s in the asymptotic regime and 8.17 plus or minus 2.53 Mbit/s using conservative finite-size parameter-estimation bounds. These results demonstrate the experimental benefits of jointly exploiting two pilot references for frequency and phase recovery in LLO CV-QKD systems, supporting their development for metropolitan quantum communication networks.
Transcript
Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.
Kai: We're looking at the paper: "Dual-Pilot-Assisted Frequency and Phase Recovery in Experimental CV-QKD".
Mira: Here's the gist of the paper. Continuous-variable quantum key distribution (CV-QKD) promises high rates and seamless integration with classical beams within a single optical fiber This work presents an end-to-end experimental CV-QKD implementation with optical transmission, coherent detection, digital signal processing, parameter estimation, information reconciliation, and privacy amplification The system achieved a final secretkey rate of approximately 5.11 Mbit/s for the asymptotic regime and 4.67 Mbit/s for the finite-size regime System Architecture and Implementation The experimental setup involves Alice using a continuous-wave (CW) laser operating at 1550 nm and an in-phase and quadrature modulator (IQM) driven by an arbitrary waveform generator (AWG) A variable optical attenuator (VOA) is placed after the IQ modulator to adjust the modulation variance of the thermal state The quantum channel consists of a 12 km single-mode fiber (SMF) with a measured average attenuation of 0.185 dB/km at 1550 nm, resulting in a total channel loss of 2.22 dB At Bob’s station, RF heterodyne detection is used to evaluate both quadratures simultaneously The two beams at the balanced beam splitter inputs are aligned in polarization with manual polarization controllers and then impinge on a balanced detector Data Processing and Security Analysis The processing subframes were independently recovered using a DSP routine based on the QOSST framework This routine first identifies the two frequency-multiplexed pilot tones to estimate carrierfrequency offset and sampling-clock mismatch After compensating for these impairments, the quantum signal was shifted to baseband, filtered using a root-raised-cosine matched filter, sampled at the optimum timing instant, and downsampled Classical post-processing includes information reconciliation using multidimensional reconciliation combined with multiedge-type low-density parity-check (MET-LDPC) errorcorrecting codes with a code rate of approximately 0.128 Following error correction, parameter estimation is performed by evaluating the Holevo bound Finally, Toeplitz-matrix privacy amplification rounds are applied to reduce Eve’s mutual information below the prescribed security level Performance Metrics and Comparison The system achieved an average modulation variance Vmod of approximately 0.92 SNU in the asymptotic regime The quantum channel exhibited untrusted transmittance of T = 0.6 and excess noise of ξB = 15.99 mSNU at Bob in the asymptotic regime In the finite-size regime, conservative confidence bounds are adopted for the channel parameters The secret key rate (SKR) is given by SKR = Rs(1 − FER)
βIAB − χBE − ∆fin(nacc): The asymptotic experimental SKR was 5.11 Mbit/s and the finite-size experimental SKR was 4.
Title and authors: Kai: So we’ve seen the technical results, but let’s start with the paper itself, Dual-Pilot-Assisted Frequency and Phase Recovery in Experimental CV-QKD. It basically describes a system that uses two independent lasers for both sides to manage those frequency drifts.
Mira: And the authors are Christiano M. S. Nascimento, Artur A. Matoso, Gustavo C. Amaral, and Guilherme P. Temporão from institutions like QuIIN in Salvador and NITeQ in Rio de Janeiro and TNO in The Netherlands.
Lev: I’m curious how their approach differs from what we've seen before with local oscillator references, which the paper says can open security loopholes for eavesdroppers if you use them improperly.
Kai: That’s the core difference, Lev. They are using fully independent transmitter and receiver lasers, which is a different way to set up the system compared to just transmitting a local oscillator reference along with the quantum channel.
Mira: The implication here is that they are trying to achieve better security by ensuring those independent sources don't accidentally create vulnerabilities for an eavesdropper who knows about the reference signal.
Lev: I wonder what this means for scaling? If we want to do this over longer distances, does the complexity of managing two independent lasers become a bigger overhead than just one well-managed local oscillator?
Kai: Well, the paper shows that for twelve kilometers, it works and gives us those secret key rates we talked about earlier. It’s a proof of concept for practical fiber deployment.
Mira: The authors are essentially demonstrating that by actively recovering frequency and phase information using these dual pilots, you can maintain a high quality link even in a noisy environment.
Lev: That suggests that the overhead of the DSP routine they developed is justified if it leads to more stable parameter estimation overall on real hardware.
Kai: It’s about making sure that when you're running this on actual superconducting qubits or whatever hardware we build, you aren't just dealing with noise, but noise you can actively compensate for.
The paper's summary: Mira: So to summarize what the paper is actually doing, they are detailing an end-to-end experimental CV-QKD implementation. It covers everything from Alice’s CW laser and the IQ modulator to Bob’s RF heterodyne detection.
Kai: And they go through all the steps: optical transmission, coherent detection, digital signal processing, parameter estimation, information reconciliation, and privacy amplification. It's a full pipeline.
Lev: I need to focus on the DSP part because that’s where most of the real engineering happens for real hardware. They describe how they use pilot tones to identify carrier frequency offset and sampling clock mismatch first.
Kai: Exactly. Then, after compensating for those issues, they shift the quantum signal to baseband, filter it with a root-raised cosine matched filter, sample it at the optimum timing instant, and downsample everything.
Mira: The classical post processing then involves information reconciliation using multidimensional reconciliation combined with multiedge type low density parity check error correcting codes.
Lev: And they use a code rate of about zero point one two eight for that error correction stage, which is what tells us how much overhead we’re adding to the raw data transmission.
Kai: After that, they perform parameter estimation by evaluating the Holevo bound and finally apply Toeplitz matrix privacy amplification rounds. It’s a complete security analysis for this link.
Mira: The results show they achieved an asymptotic secret key rate of about five point one one megabits per second and a finite-size regime rate of four point six seven megabits per second.
Lev: Those numbers are the output we need to look at when deciding if this is something that can actually be run on current error correcting hardware without needing massive overhead.
Kai: It’s about showing that you can get a usable key rate even with all these steps in place, which is the main goal of the experimental work.
The paper's improvements: Lev: The paper suggests several improvements, and I’m looking at how those suggestions translate into practical engineering changes. One improvement they point to is using puncturing and shortening to dynamically adjust the effective code rate based on measured signal-to-noise ratio.
Kai: That sounds like a smart way to optimize reconciliation efficiency when the noise level changes during transmission, which makes sense if you're not in a perfectly stable environment.
Mira: I agree that dynamic adjustment is powerful, because it allows you to balance the trade-off between how efficiently you reconcile information and how often you have to perform decoding.
Lev: From a hardware standpoint, that means the error correction code rate isn't fixed; it changes based on what the signal actually looks like in real-time.
Kai: And another thing they mention is channel characterization stability. They show that their estimated parameters remained statistically stable, meaning those asymptotic point estimates were close to finite-size confidence bounds.
Mira: That’s good because if the parameters drift wildly, you can't trust your security analysis or your key rate calculation.
Lev: So they’re saying that the method for estimating the channel characteristics is robust enough that even when you have fewer symbols, those estimates are reliable enough to build a secure key on.
Kai: It means the method for tracking impairments is more stable than we might have thought, which is encouraging for experimentalists trying to characterize new optical fiber types.
Conclusion: Mira: So wrapping up the Dual-Pilot-Assisted Frequency and Phase Recovery in Experimental CV-QKD paper, the main implication is that this dual pilot assistance helps stabilize the link parameters under realistic conditions.
Kai: It gives us a concrete experimental baseline for metropolitan fiber deployment, showing we can achieve secret key rates around five point one one megabits per second asymptotically.
Lev: For me, it’s about confirming that the overhead of the QOSST framework is manageable when you need those stable parameter estimates to run on actual hardware.
Mira: And they're also suggesting adaptive information reconciliation for low-SNR regimes, which means we can optimize the code rate dynamically based on what we measure.
Kai: So it’s about building a system that can handle the noise and still produce a usable key rate, even with the limitations of finite-size experiments.
Lev: Exactly. The paper shows how to keep things running reliably when you have to make those trade-offs between speed and security on real hardware.
Mira: That’s it for this paper, Dual-Pilot-Assisted Frequency and Phase Recovery in Experimental CV-QKD, demonstrating a solid path forward for practical CV-QKD implementation.
More episodes
- 2603.29754-Coherent and dissipative energy transfer in driven open quantum systems: a driven quantum master equation approach
- 2604.01157-Thermal Entanglement and Out-of-Equilibrium Thermodynamics in 1D Bose Gases
- 2604.06410-Directional and correlated optical emission from a waveguide-engineered molecule with local control
- 2604.07214-Quantum Gibbs sampling through the detectability lemma
- 2604.13165-Finite-size Entanglement Certification via Third-Order Local Randomized Measurements
- 2604.18542-Dissipative Preparation of Correlated Quantum States in Dipolar Rydberg Arrays
- 2605.00096-Scalable spin-nematic squeezing in multi-level dipole-interacting Rydberg atom arrays
- 2605.02565-Sample-Based Quantum Diagonalization with Amplitude Amplification
- 2605.20111-Mechanism of wavefunction collapse in measurements of separated quantum subsystems
- 2606.05097-No-Go Theorem for Gaussian Quantum Repeaters from Fractional Extendibility