REVIEW 2 major objections 5 minor
Co-transmission of classical data and continuous-variable entanglement over a single quantum channel
T0 review · 2 major / 5 minor · reviewed 2026-08-01 · deepseek-v4-flash
Pith's one-line read The paper shows that a single lossy optical mode can carry both classical bits and continuous-variable Gaussian entanglement, with positive distillable entanglement when classical errors stay below a loss-dependent threshold.
desk verdict Good hybrid classical-quantum protocol analysis, but the RCI/one-way mismatch in footnote 58 needs to be resolved before the entanglement-distribution claim holds. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The key mechanism is the dual use of continuous-variable teleportation as a demodulator: Bob mixes Alice's received mode with his own EPR state, and the dual-homodyne measurement outcomes both teleport the quantum state (via displacement feed-forward) and estimate Alice's classical QPSK symbol. The load-bearing identity is the output covariance matrix V_out = [[a0 I, c0(1−δ)σZ], [c0(1−δ)σZ, (b0+Δ)I]] with Δ = 2α²τ e_C, which quantifies how classical bit errors convert into an effective excess-noise term on the entangled state. The paper then invokes Gaussian extremality to turn this Gaussian surrogate into lower bounds on RCI, entanglement of formation, and the Devetak-Winter key rate.
What would settle it
Compute or measure the one-way distillable entanglement of the output state for parameters where the paper's RCI bound is positive (e.g., η=30 dB loss, ε=0.025 SNU, e_C=10^-6); if the forward-only distillable entanglement is zero while the RCI lower bound is positive, the claim that the protocol distributes distillable entanglement under one-way communication is falsified. Alternatively, an experimental attempt to distill EPR entanglement from the output state under one-way classical communication at 30 dB loss with e_C=10^-6 would settle it.
Extended reading notes
Core claim
The central claim is that the CMQC protocol distributes two-mode-squeezed (Gaussian) entanglement and classical information on the same optical mode. The decoupling trick is a continuous-variable teleportation performed by Bob: his homodyne outcomes are a noisy estimate of Alice's classical displacement, so he can recover the classical message while the entanglement is transferred to his retained mode. Classical bit errors make the final state non-Gaussian, but the contamination is exponentially small in the classical signal strength, and the ensemble state is captured by the covariance matrix V_out = [[a0 I, c0(1−δ)σZ], [c0(1−δ)σZ, (b0+Δ)I]], where Δ = 2α²τ e_C is the extra noise due to mis
Load-bearing premise
The central bound on distillable entanglement is computed with the reverse coherent information, which assumes Bob may help via a backward classical channel, while the protocol is presented as one-way; the paper explicitly brackets this contradiction 'for practical reasons.'
Editorial extensions
If this is right
- Existing classical optical links could be upgraded to distribute CV entanglement while keeping the classical data stream, by adding an entanglement source and a teleporter at each node.
- The protocol's classical bit-error rate can be tuned to a target (e.g., 10^-9) by increasing the modulation amplitude, at the cost of a large extra noise term 2d² e_C that kills entanglement beyond a loss threshold.
- The point-to-point secret-key rate saturates the optimized CV-QKD bound only at very low error rates (e_C ≤ 10^-14), making the entanglement-based scheme more fragile to classical errors than the earlier SQCC-QKD approach.
- The scheme naturally extends to repeater-like configurations by adding noiseless linear amplification and error-corrected re-displacement, allowing the hybrid signal to propagate over longer distances.
Reading between the lines
- One consequence left implicit is that a network of classical transceivers could maintain a background reservoir of entanglement with every other node, enabling on-demand QKD, teleportation, or distributed sensing without a dedicated quantum channel.
- The use of reverse coherent information under a one-way protocol is the paper's most fragile step; replacing RCI with the forward-assisted capacity would either require two-way communication in the protocol or could substantially lower the predicted distillable-entanglement thresholds.
- Gaussian extremality is invoked for a non-Gaussian mixture; a direct computation of the distillable entanglement of the exact non-Gaussian state (or a counterexample) would determine whether the lower bound is tight.
- The sharp transition between distillable and non-distillable regimes as a function of e_C suggests a practical engineering rule: the classical signal-to-noise ratio at Bob's receiver must exceed a threshold set by the tolerable excess noise of the quantum layer.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript theoretically analyzes a protocol that aims to transmit classical data and continuous-variable (CV) Gaussian entanglement simultaneously on a single optical mode. Alice prepares a two-mode squeezed vacuum, classically modulates one mode via QPSK displacements, and sends it through a lossy thermal channel. Bob performs a CV teleportation of the received mode using his own two-mode squeezed state; the dual-homodyne outcomes provide both the classical demodulation and the feedforward correction needed to approximate a zero-mean entangled state. The authors derive the bit-error rate, the ensemble covariance matrix of the output state, and lower bounds on reverse coherent information, entanglement of formation, and asymptotic secret-key rate. They conclude that positive reverse coherent information is achievable for sufficiently low classical bit-error rates, e.g. e_C ≤ ~5e-7 at 30 dB loss, and compare the scheme with the SQCC-QKD protocol of Ref. [27].
Significance. If the central claim is correct, the protocol is a conceptually useful step toward integrating classical communication with entanglement distribution in future quantum networks: existing classical links could, in principle, be augmented to distribute CV entanglement without sacrificing classical service. The analytic derivation of the output covariance matrix and the bit-error rate is detailed and, on its face, internally consistent; the paper also gives explicit formulas for RCI, E_F, and K∞ that can be checked by others. These are genuine strengths. However, the main quantitative evidence for "distillable entanglement" rests on an assumption about the operational meaning of the reverse coherent information that is explicitly acknowledged as contradictory in footnote 58, and the Gaussian-extremality step for RCI is asserted rather than proved. These issues directly affect the headline threshold numbers and must be resolved before the central claim is established.
major comments (2)
- [Section IV and Appendix B, Eq. (43)/(B1)–(B4); footnote 58] The protocol is described as one-way: Alice transmits, Bob teleports and applies local feedforward, and no backward classical channel is used. Yet the distillable-entanglement claim is quantified by the reverse coherent information R = S(ρ_A) − S(ρ_AB), which is a lower bound on reverse-assisted or two-way entanglement distillation, not on one-way distillable entanglement. The manuscript never reports the forward coherent information I(A>B) = S(ρ_B) − S(ρ_AB), which would be the standard one-way lower bound. Footnote 58 openly acknowledges: "It may seem unreasonable to claim that Bob does not have a backwards communications channel... We are obliged to ignore this apparent contradiction." This is not merely cosmetic: if R > 0 but I(A>B) ≤ 0 in the plotted regime, the stated one-way protocol does not distribute one-way distillable entanglement according to the presented evidence. Please e
- [Appendix B, Eq. (B4)] The lower bound R_out ≥ S(ρ_G_A) − S(ρ_G_AB), where ρ_G is the Gaussian state with the same covariance matrix as the non-Gaussian ensemble output, is attributed to "Gaussian extremality" without proof or a specific theorem reference. The output is a mixture of Gaussian components with different displacement means, and replacing it by a single zero-mean Gaussian is nontrivial for the difference S(A) − S(AB): Gaussian states maximize the entropies S(A) and S(AB) separately for fixed covariance, but it is not automatic that the difference is minimized by the Gaussian state. If this extremality step fails, the RCI thresholds in Figs. 2(a) and 3 are not established. Please supply a proof or a precise citation for this particular use of Gaussian extremality.
minor comments (5)
- [Appendix A, Eq. (A27)] The four branches of the conditional displacement are all labeled "x0 ≥ 0, y0 ≥ 0"; they should be the four sign combinations. This is clearly a typo, but it makes the appendix hard to follow.
- [Eqs. (26)–(27) vs. (37)–(38)] The relation between the single-symbol covariance matrix, which contains the term −||µ_B^{d1}||², and the full-alphabet covariance matrix, which does not, is implicit. A brief explanation that the full-alphabet covariance combines the average of the component covariances with the covariance of the component means would remove ambiguity.
- [Throughout] The protocol is called "Gaussian CMQC" although the output state is explicitly non-Gaussian and only approximated by a Gaussian state. A more precise term, such as "Gaussian-approximated CMQC," would avoid confusion.
- [Section IV, key-rate calculation] The key-rate lower bound assumes Alice and Bob perform a heterodyne measurement on the output state, while the only measurement described in the protocol is Bob's dual-homodyne teleportation measurement. If additional measurements on the retained modes are intended for key generation, this should be stated explicitly and reconciled with the protocol timing.
- [Figure 3] The color scale in the heat maps is not described adequately for the printed version; please add a clear caption describing the shading and the boundary line.
Circularity Check
No circular reduction found; analytic derivation is self-contained, though it leans on prior self-citations for protocol structure and on RCI with an admitted reverse-assistance mismatch.
full rationale
The output state V_out is derived analytically from explicit Gaussian operations (Appendix A), with the classical bit-error rate e_C and the correction terms δ, Δ obtained by direct quadrature integration (Eqs. 18-21, A42-A43); no parameter is fitted to the target entanglement or key-rate quantities. R, E_F, and K_∞ are evaluated from V_out via symplectic eigenvalues and standard extremality arguments, not by construction equal to any input. The QoS condition Eq. (45) fixes d for a chosen e_C; this is a design constraint, not a hidden fit. Citations to [1] and [27] supply the protocol architecture and the threshold-discrimination rule, but the paper independently characterizes the resulting state; importing a protocol design from prior work is not a circular derivation. Footnote 58 explicitly concedes that quantifying one-way distillable entanglement by reverse coherent information is inconsistent; that is a substantive correctness limitation, not a circular reduction. The self-citations (Ralph in [33,34,39,40,57]) are to standard methods or numerical tools and are not used as the sole evidence for the central result.
Assumptions & free parameters
free parameters (2)
- r_A, r_B (squeezing parameters) =
optimized over [0,∞) (unconstrained) or [0,1.151] (10 dB constrained)
- classical signal amplitude d (or α) =
set by Eq. (45) for target e_C ∈ {10^-9, 10^-6}
assumptions (5)
- domain assumption Gaussian extremality: for fixed covariance matrix, Gaussian states minimize entropy and entanglement measures, so RCI/EoF/key-rate bounds can be computed from the Gaussian state with V_out.
- domain assumption Optimality of Gaussian attacks: Eve's Holevo information is upper-bounded by the Gaussian-purified state (refs [61,62]).
- domain assumption CV teleportation with gain g_x=-√2 tanh r_B is equivalent to a pure-loss channel with transmissivity τ=tanh^2 r_B.
- domain assumption Channel is a beamsplitter with transmissivity η and thermal noise n=ηε/[2(1−η)] (excess noise ε in SNU).
- ad hoc to paper Reverse coherent information lower-bounds one-way distillable entanglement.
Cite this review
Pith. "Pith review of Co-transmission of classical data and continuous-variable entanglement over a single quantum channel." pith.science (2026). https://pith.science/paper/DGPQSZW4
@misc{pith2026260725179,
author = {Pith},
title = {Pith review of: Co-transmission of classical data and continuous-variable entanglement over a single quantum channel},
year = {2026},
howpublished = {\url{https://pith.science/paper/DGPQSZW4}},
note = {Machine review of arXiv:2607.25179}
}
read the original abstract
Displacement-based simultaneous quantum-classical communications (SQCC) protocols, as originally proposed, are generally incompatible with the majority of useful quantum communication schemes, such as entanglement distribution or repeater-based quantum key distribution: direct measurement of the classical signal also measures and destroys the quantum state, leaving point-to-point Gaussian quantum key distribution as the only quantum communication scheme amenable to integration with SQCC. In this work, we apply the classically-modulated quantum communication protocol proposed by Zaunders and Ralph [arXiv:2606.03181v3] to circumvent this issue and demonstrate the distribution of continuous-variable Gaussian entanglement simultaneously with classical information. We characterise the quality of the distributed entangled state and outline how the scheme is suitable for use in repeater-based networks. Lastly, we compute the secret key generation rate of the Gaussian CMQC scheme in the point-to-point case and compare it to the equivalent point-to-point SQCC protocol.
Figures
Figures from the paper (1 more)
Reviewed August 1, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.