Finite-energy GKP-QPC architectures for photonic quantum memories and repeaters

arXiv:2608.15620 · quant-ph · Submitted 2026-08-16 · 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: We're looking at the paper: "Finite-energy GKP-QPC architectures for photonic quantum memories and repeaters".

Mira: Here's the gist of the paper. Finite-energy GKP-QPC architectures for photonic quantum memories and repeaters study a concatenated architecture in which each physical rail is a finitely squeezed GKP qubit transmitted through a pure-loss fiber segment, corrected by teleportation-based GKP error correction with finitely squeezed ancillae, and decoded by an outer quantum parity code (QPC) Concatenated Architecture Framework The central idea of this work is therefore to use the GKP layer to convert continuous-variable loss noise into effective rail-level Pauli error statistics, and then use the QPC layer to suppress the remaining logical errors The elementary module studied in this work is illustrated schematically in Fig.

Kai: 1 This gives a common finite-squeezing framework for both all-optical quantum memories and one-way quantum repeaters without pre-amplification Thresholds and Fidelity in Memory At zero propagation loss, where finite GKP ancilla squeezing is the only noise source, we find a finite-squeezing threshold of 5.06 dB for the concatenated GKP–QPC architecture In the memory setting, the QPC layer lowers the squeezing at which repeated correction becomes useful from 6.7 dB for bare GKP correction to 5.2 dB for QPC(3, 3) and 4.3 dB for QPC(5, 5) The relative fidelity improvement is largest in the intermediate-noise regime, where the GKP layer is sufficiently accurate to leave mostly correctable rail errors but not so accurate that the outer code becomes redundant Repeater Performance and Optimization In the repeater setting, avoiding pre-amplification gives higher secret-key fractions at moderate squeezing, which is experimentally relevant because high-quality GKP squeezing remains a major bottleneck The advantage is not monotonic: at high squeezing, loss-induced peak shifting reduces performance and produces an optimal squeezing Optimizing over QPC size reveals interior optima, rather than a monotonic preference for larger codes Design Rules and Tradeoffs The results provide quantitative design rules for finite-squeezing GKP–QPC quantum memories and repeaters The optimal segment length Lopt = Ltot/Nopt EC for a given ε is found by balancing photon loss accumulation against the noise introduced by each teleportation-based correction step QPC concatenation should be used in the clean-hop regime where the inner GKP layer has already reduced the noise to a correctable level The outer QPC layer is most valuable when the elementary GKP-corrected rails are already sufficiently reliable, allowing QPC decoding to convert moderate physical reliability into a large logical advantage Outside this regime, the optical-mode overhead outweighs the logical error suppression The optimal code is not at the largest tested code but at an interior point, QPC(7, 5), yielding a per-mode SKF of approximately 2.313 × 10−2 Final Summary Overall, these results show that finite-squeezing GKP–QPC concatenation can substantially improve photonic quantum memories and one-way repeaters, but only within a sharply regime-dependent operating window The outer QPC layer is most valuable when the elementary GKP-corrected rails are already sufficiently reliable, allowing QPC decoding to convert moderate physical reliability into a large logical advantage The optimal code is not at the largest tested code but at an interior point, QPC(7, 5), yielding a per-mode SKF of approximately 2.

Title and authors: Kai: So, we're looking at this paper today titled "Finite-energy GKP-QPC architectures for photonic quantum memories and repeaters". It sounds pretty technical, but it's about how to build these quantum systems that actually work with the real stuff we have—finite energy states and pure loss fiber transmission.

Mira: Yeah, the authors are Kaustav Chatterjee and Ulrik Lund Andersen from the Center for Macroscopic Quantum States at TU Denmark. They’re diving into a concatenated architecture where you take a physical rail, squeeze it with GKP states, send it through lossy fiber, correct that with teleportation based GKP error correction using finitely squeezed ancillae, and then decode everything with an outer quantum parity code.

Lev: It makes sense because we're always dealing with noise sources like loss and measurement errors in hardware. The authors are proposing this structure as a common framework for both all-optical quantum memories and one-way repeaters without needing any preamplification, which is pretty appealing for experimental setups.

Kai: Exactly. I'm interested in how they've managed to model the noise introduced by that pure loss fiber segment—is it just treating it as some kind of random displacement channel?

Mira: Well, page two explains that when you have loss, there are two effects: first, the variance of those Gaussian peaks increases because of the vacuum noise added; and second, each individual peak gets displaced toward the origin in phase space. These two things together cause logical errors by pushing probability distributions into wrong bins.

Lev: That sounds like a real challenge for any continuous variable system we build. So they’re not ignoring that the squeezing level matters—that's a key part of what they're exploring here, right?

The paper's summary: Kai: Okay, so the main idea is using GKP states to convert those continuous homodyne syndromes into effective rail-level Pauli marginals. Then the outer quantum parity code takes those marginals and suppresses the residual qubit-level errors. It’s a layered approach to error correction.

Mira: Right, so if you think of it this way, the GKP layer is doing its job to make the noise look like rail errors that are easier to handle with a standard QPC. The QPC layer then cleans up what’s left over from the process.

Lev: That layered approach is interesting because it tackles the noise at different levels of abstraction. The paper shows how this same elementary module can be repeated in time to make a memory, or concatenated in space to make a repeater, which gives it versatility for both applications.

Kai: I saw they mention avoiding preamplification. That's important because if you amplify lossy states, you might obscure the finite energy peak-shifting effect that is intrinsic to pure loss itself.

Mira: Exactly. Page two points out that at low GKP squeezing, the dominant error mechanism is actually the broadening of those GKP peaks, but as squeezing gets higher, they become more sensitive to loss-induced inward displacement in phase space. It’s a trade-off we need to watch for.

Lev: That competition between peak broadening and displacement seems like a core physical constraint they are working within. So, the architecture is designed to work best in that intermediate noise regime where the GKP layer is accurate enough for rail errors but not so accurate that the outer code becomes useless.

The paper's improvements: Kai: The paper gives some concrete numbers on thresholds, which is always good. They find a finite-squeezing threshold of five point zero six decibels for this concatenated GKP-QPC architecture when propagation loss is zero. That's the baseline.

Mira: And then they show how the QPC layer actually helps lower that required squeezing level for repeated correction to be useful. For bare GKP correction, you need six point seven decibels of squeezing, but with a QPC of three by three, you only need five point two decibels.

Lev: That's a significant reduction in the required resource level for the outer code. And they show that for a QPC of five by five, the threshold drops even further down to four point three decibels. That suggests the benefit of concatenation scales with how robust your outer code is designed.

Kai: I wonder what this means practically? It tells us that you can use weaker initial squeezing and still get good error suppression if you layer on a strong enough QPC, like a five by five one.

Mira: The relative fidelity improvement they found is largest in the intermediate-noise regime. That's where the GKP layer is accurate enough to leave mostly correctable rail errors, but not so accurate that the outer code becomes redundant. It’s a sweet spot for performance gains.

Lev: So, what about repeaters? They say avoiding preamplification gives you higher secret-key fractions at moderate squeezing, which is relevant because high-quality GKP squeezing is still a major experimental bottleneck for us.

Conclusion: Kai: So wrapping this up on the "Finite-energy GKP-QPC architectures for photonic quantum memories and repeaters," the main thing they show is that concatenation can substantially improve both photonic quantum memories and one-way repeaters when you're dealing with finite GKP squeezing and pure loss.

Mira: The most important implication, which I think everyone should get, is that the outer QPC layer isn't useful if the inner GKP correction doesn't already make the rails reliable enough to convert physical reliability into a large logical advantage.

Lev: I agree with Mira there. They also show that they didn't find that larger QPC codes are always better; they found interior optima, specifically at the QPC size of seven by five, which gave a per-mode secret key fraction of approximately two point three times ten to the minus two.

Kai: So to summarize for the listener, this architecture works best when you have enough GKP correction to leave manageable rail errors that a QPC can handle effectively.

Mira: That’s right, and they found that the optimal code size isn't necessarily the biggest one they tested, but rather an interior point like seven by five.

Lev: For me, what this means is that if you're building this for real hardware, you need to balance photon loss accumulation against the noise from each teleportation step to find your best segment length L opt over N opt EC.

Kai: Yeah, and they gave us a design rule: use the QPC concatenation in the clean-hop regime where that inner GKP layer has already reduced the noise to a correctable level. That’s what we need to focus on for our next experiments.

Kaustav Chatterjee, *, Ulrik Lund Andersen, ^

Center for Macroscopic Quantum States (bigQ) · Department of Physics, Technical University of Denmark

quant-ph

Submitted: 2026-08-16

Updated: 2026-11-09

Comments: 15 pages, 12 figures

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 87/100

The gist: Finite-energy GKP-QPC architectures for photonic quantum memories and repeaters study a concatenated architecture in which each physical rail is a finitely squeezed GKP qubit transmitted through a…

Key concepts

Finite-squeezing GKP qubit
This is a type of quantum state used as a qubit that has limited squeezing. It's employed in this architecture to convert continuous-variable loss noise from optical fibers into discrete, manageable Pauli errors at the rail level. This allows for error correction without needing perfect initial squeezing.
Concatenated Architecture Framework
The system uses two layers of error correction: an inner GKP layer and an outer QPC layer. The GKP layer handles the noise introduced by loss during transmission, converting it into errors that the outer QPC can then correct. This layered approach is key to achieving robust quantum memory and repeater performance.
QPC(n, m) Code
This refers to a specific type of Quantum Parity Code used as the outer layer for error correction. The paper tests different sizes, like QPC(3, 3) and QPC(5, 5), to determine which code size provides the best performance (highest secret key fraction) under varying noise conditions.
Regime-dependent Concatenation Utility
The effectiveness of using the outer QPC layer depends on the reliability achieved by the inner GKP layer. The paper finds that concatenation is most valuable when GKP correction has made the rails 'sufficiently reliable,' allowing the QPC to convert moderate physical errors into a significant logical advantage.

Terminology

Summary

Finite-energy GKP-QPC architectures for photonic quantum memories and repeaters study a concatenated architecture in which each physical rail is a finitely squeezed GKP qubit transmitted through a pure-loss fiber segment, corrected by teleportation-based GKP error correction with finitely squeezed ancillae, and decoded by an outer quantum parity code (QPC)

Concatenated Architecture Framework

The central idea of this work is therefore to use the GKP layer to convert continuous-variable loss noise into effective rail-level Pauli error statistics, and then use the QPC layer to suppress the remaining logical errors The elementary module studied in this work is illustrated schematically in Fig. 1 This gives a common finite-squeezing framework for both all-optical quantum memories and one-way quantum repeaters without pre-amplification

Thresholds and Fidelity in Memory

At zero propagation loss, where finite GKP ancilla squeezing is the only noise source, we find a finite-squeezing threshold of 5.06 dB for the concatenated GKP–QPC architecture In the memory setting, the QPC layer lowers the squeezing at which repeated correction becomes useful from 6. p. 17 dB for bare GKP correction to 5. p. 02 dB for QPC(3, 3) and 4. p. 03 dB for QPC(5, 5) The relative fidelity improvement is largest in the intermediate-noise regime, where the GKP layer is sufficiently accurate to leave mostly correctable rail errors but not so accurate that the outer code becomes redundant

Repeater Performance and Optimization

In the repeater setting, avoiding pre-amplification gives higher secret-key fractions at moderate squeezing, which is experimentally relevant because high-quality GKP squeezing remains a major bottleneck The advantage is not monotonic: at high squeezing, loss-induced peak shifting reduces performance and produces an optimal squeezing Optimizing over QPC size reveals interior optima, rather than a monotonic preference for larger codes

Design Rules and Tradeoffs

The results provide quantitative design rules for finite-squeezing GKP–QPC quantum memories and repeaters The optimal segment length Lopt = Ltot/Nopt EC for a given ε is found by balancing photon loss accumulation against the noise introduced by each teleportation-based correction step QPC concatenation should be used in the clean-hop regime where the inner GKP layer has already reduced the noise to a correctable level

Conclusion on Concatenation Utility

The outer QPC layer is most valuable when the elementary GKP-corrected rails are already sufficiently reliable, allowing QPC decoding to convert moderate physical reliability into a large logical advantage Outside this regime, the optical-mode overhead outweighs the logical error suppression The optimal code is not at the largest tested code but at an interior point, QPC(7, 5), yielding a per-mode SKF of approximately 2.313 × 10−2

Final Summary

Overall, these results show that finite-squeezing GKP–QPC concatenation can substantially improve photonic quantum memories and one-way repeaters, but only within a sharply regime-dependent operating window The outer QPC layer is most valuable when the elementary GKP-corrected rails are already sufficiently reliable, allowing QPC decoding to convert moderate physical reliability into a large logical advantage The optimal code is not at the largest tested code but at an interior point, QPC(7, 5), yielding a per-mode SKF of approximately 2.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Finite-squeezing GKP–QPC concatenation can be used to substantially improve photonic quantum memories and one-way repeaters under finite GKP squeezing and pure-loss transmission by converting each lossy optical rail into an effective qubit-level channel, while the outer QPC layer suppresses the residual rail-level errors.

  2. The improved system can determine the optimal code size for a given total distance and repeater spacing by scanning over QPC parameters to maximize per-mode secret key fraction, such as finding QPC(7, 5) as an interior optimum for a 104 km link.

  3. The AI system can predict the regime where concatenation is most beneficial by identifying that the outer code is most valuable when the elementary GKP-corrected rails are already sufficiently reliable, allowing QPC decoding to convert moderate physical reliability into a large logical advantage.

Sources

Related papers