{"id":"37389afa-6bf6-44f3-87a1-1241e6c8a4e8","arxiv_id":"2607.26905","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A fourth-degree polynomial spectral profile, scaled by source power and effective fiber length, fits measured C-band SpRS from 1310 nm commercial SFPs and identifies a noise minimum near 1535 nm.","lead":"Researchers measured Raman noise that O-band classical light dumps into the C-band in ordinary fiber, using commercial transceiver modules, and fit a simple source-scaled model of that noise versus wavelength and length. The fit points to a quieter C-band window near 1535 nm for quantum signals sharing urban fiber with 1310 nm classical traffic.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"The claimed source-independent, transferable A(λ) rests on a single empirical fourth-degree fit whose coefficients and uncertainties are unpublished, so the channel-optimization claim cannot be independently checked or applied.","rationale":"The Reader correctly isolates the weakest link: the single empirical A(λ) plus fixed attenuations is asked to carry source-independence and applicability beyond the measured spools. My stress-test simply sharpens the same point—without the numerical coefficients the fit is non-reproducible, so the channel-optimization recommendation and the “parameter-robust” language cannot be verified. No deeper inconsistency in the physics or experimental design appears; the qualitative minimum near 1535 nm is visible in the raw counts and is useful. Hence the verdict remains CONDITIONAL, exactly as the Reader concluded: accept the experimental guidance once coefficients/uncertainties are released and generality claims are narrowed to the characterized regime. No stronger objection is warranted.","tokens_in":9043,"tokens_out":613,"duration_ms":11988,"concrete_test":"Publish the four polynomial coefficients of A(λ) together with their standard errors (or the full covariance matrix) and the exact α_o, α_c values used. An independent group can then recompute C_SpRS on any of the published length/power combinations; if the reconstructed curve deviates from the paper’s figures by more than the tabulated relative errors, or if the location of the minimum shifts by more than one ITU channel, the transferability claim fails.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim is that C_SpRS(S_src,λ,L)=S_src A(λ) L_eff with a single fourth-degree polynomial A(λ) (fixed at λ_o=1310 nm) is source-agnostic and sufficiently parameter-robust to identify the optimal C-band channel (~1535 nm / ch. 44) for diverse fiber systems. That claim stands or falls on A(λ) itself. Section V states that A(λ) is obtained by polynomial regression on the combined spool data and that degree 4 is preferred; Figs. 4–7 and Table I then show relative errors that already reach 14–32 % on the shortest (100 m) spools and remain several percent even at 5 km. No numerical coefficients, covariance, or residual spectrum for the polynomial are supplied, nor is any cross-validation against an independent fiber type or a different pump wavelength performed. Consequently the asserted minimum, the quantitative noise estimates, and the generality language in the abstract and §VI cannot be reproduced or stress-tested by a reader; the model remains a private descriptive fit rather than a transferable predictive tool.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The manuscript reports experimental measurements of C-band spontaneous Raman scattering (SpRS) generated by co-propagating O-band classical traffic at 1310 nm in standard single-mode fiber, using commercial SFP transceivers as well as a narrow-linewidth laser. Cascaded WDMs and a tunable narrowband filter isolate SpRS counts across ITU channels 15–63 for fiber lengths of 100 m, 500 m and 5 km. From these data the authors propose the compact model C_SpRS(S_src, λ, L) = S_src A(λ) L_eff (Eq. 3), where L_eff is the usual effective length and A(λ) is a single fourth-degree polynomial fitted to the combined spool measurements. They report a spectral minimum near ~1535 nm (channel 44), relative errors in Table I, and argue that the model is source-independent and therefore usable to allocate optimal C-band quantum channels in the presence of O-band classical traffic.","tokens_in":9367,"tokens_out":1428,"duration_ms":33580,"significance":"Quantum–classical coexistence on deployed single-core fiber is a genuine deployment bottleneck; measurements that employ commercial SFPs rather than laboratory lasers, and that scan the full C-band DWDM grid, are practically valuable. The consistent observation of a count minimum near 1535 nm across three sources and several lengths is a concrete, actionable finding for the authors’ testbed and similar urban links. If the claimed source-agnostic, length-and-power-scalable A(λ) were fully specified and shown to transfer, the work would supply a useful engineering tool. At present the contribution is best read as a careful empirical characterization plus a descriptive fit, not yet as a parameter-robust predictive model for diverse systems.","major_comments":[{"comment":"§V and Eq. (3): A(λ) is defined as a fourth-degree polynomial “derived from the experimental measurements,” yet no coefficients, covariance, residual spectrum, or fitting procedure (weights, wavelength grid, joint vs. sequential fit) are supplied. Without these numbers the model cannot be reproduced, independently checked, or applied by a reader to a new link. This is load-bearing for every claim of a “predictive,” “compact,” and “parameter-robust” tool in the abstract and §VI.","section":"§V, Eq. (3)"},{"comment":"§V, Figs. 4–7 and Table I: Validation largely re-compares the same (or power-normalized) spool datasets used to build A(λ). Relative errors already reach 14–32 % on 100 m spools and remain several percent at 5 km (e.g., 13.3 % commercial source, 5 km). The manuscript acknowledges that short-fiber Raman is weak relative to background, but still presents those points as supporting “robustness.” A held-out fiber length, a different fiber type, or an independent pump wavelength would be needed to substantiate transferability; none is provided.","section":"§V, Table I"},{"comment":"§VI and abstract: The language “parameter-robust description \to applicable to diverse fiber-based systems” and “independent of the specific optical source” overreaches the evidence. All data use a single pump wavelength (1310 nm), standard SMF spools (plus a brief mention of a 7 km urban loop without quantitative comparison), and three sources whose differences are absorbed into a free scale S_src. Fixed α_o = 0.35 dB/km and α_c = 0.2 dB/km are stated without measured values or sensitivity analysis. The generality claim should be narrowed to the measured regime, or additional cross-checks added.","section":"§VI, Abstract"},{"comment":"Fig. 8 / §VI: The comparison with “the theoretical model” is central to the claim that the empirical fit captures structure missed by established Raman theory, yet the theoretical curve’s precise inputs (Raman gain spectrum used, filter bandwidth, absolute scaling) are not stated. Without that specification the reader cannot judge whether the missing minimum is a genuine fiber/source effect or a mismatch in normalization or spectral resolution.","section":"§VI, Fig. 8"}],"minor_comments":[{"comment":"Eq. (1) writes β(λ_o λ_c) without a comma or semicolon; later text uses β(λ_o, λ_c). Standardize notation.","section":"§II, Eq. (1)"},{"comment":"α_o and α_c are given in dB/km in the text of §V but enter Eq. (2)–(3) as neperian coefficients; state the unit conversion explicitly.","section":"§V"},{"comment":"Fig. 3 caption and body refer interchangeably to “WDM” and “DWDM” for the C-band cascade; clarify which devices are used where.","section":"§IV, Fig. 3"},{"comment":"The 7 km urban-loop validation is asserted in §VI with a citation to an invited workshop paper but no counts, errors, or overlay figure appear here; either add a brief quantitative panel or soften the claim.","section":"§VI"},{"comment":"Typos / style: “WMD” for WDM (p. 4); “e −αoL” spacing in Eq. (2); “National Quantum Internet.it” formatting; duplicate reference [15]/[18].","section":null},{"comment":"State the SNSPD detection efficiency and DCR at the actual scanned wavelengths (not only 1550 nm) or confirm that the relative spectral shape is unaffected.","section":"§III"}],"recommendation":"major_revision","confidential_remarks":"The experimental campaign is competent and the channel-44 minimum is useful for the authors’ testbed. The main risk is over-selling an unpublished empirical polynomial as a general predictive model. If the authors release the A(λ) coefficients, tighten the generality language, and add at least one held-out check, the paper becomes a solid incremental contribution; without that it remains a local characterization. Scope is appropriate for a letters/short-paper venue in quantum networking or applied optics; full generality claims would need a longer format and more fibers/pumps."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The usable core here is experimental, not theoretical. They measure C-band SpRS from 1310 nm commercial SFPs (plus a lab laser) on standard SMF spools up to 5 km, with careful WDM cascading, and consistently see a spectral minimum around ~1535 nm / ch. 44. That opposite wavelength allocation (O classical, C quantum) and the SFP realism are the parts that matter for metro hybrid links and their Naples testbed. The filtering study and multi-source checks are done carefully enough that the qualitative channel-guidance result looks real.\n\nWhat is not new is the scaffolding. Eq. (1)–(2) are standard SpRS / L_eff. Eq. (3) just multiplies a per-source scale S_src by an empirical fourth-degree A(λ) fit to their own counts. Degree-4 is chosen by eye against the same family of curves; relative errors in Table I hit 20–30% on 100 m and stay several percent even at 5 km. No polynomial coefficients, uncertainties, or residual spectrum are published, so a reader cannot reproduce the quantitative noise estimates or stress-test the claimed minimum without re-fitting. The abstract and §VI language about a parameter-robust, source-agnostic description for “diverse fiber-based systems” overreaches what three sources at fixed λ_o on spool fiber can support. Short-fiber Raman is weak versus background; they note it, but still fold those points into the same A(λ).\n\nMath and citations are fine for this genre: L_eff is used correctly, α values are conventional, and the Raman/QKD coexistence literature is cited in the right places. Circularity is real but ordinary for an empirical spectral fit—treat A(λ) as descriptive of this setup, not as a transferable Raman coefficient.\n\nWho gets value: people building O/C coexistence on urban SMF who need a measured channel map, not a first-principles Raman theory paper. I would send it to peer review as a preliminary experimental note; require the fit coefficients, error bars, and narrower generality claims. Worth engaging if you care about practical channel selection; skip if you wanted a new physical model of SpRS.","headline":"Useful spool measurements and a practical C-band minimum near 1535 nm for O-band classical; the “source-independent predictive model” is mostly a private polynomial fit plus standard L_eff.","tokens_in":10016,"tokens_out":563,"would_cite":true,"duration_ms":23763,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"A source-independent model of Raman noise picks the quietest C-band channels for quantum light sharing fiber with O-band classical traffic.","keywords":["quantum-classical coexistence","spontaneous Raman scattering","C-band quantum channels","O-band classical traffic","single-mode fiber","SFP transceiver","DWDM channel allocation","urban fiber networks"],"falsifier":"Measure SpRS counts versus C-band wavelength on a different fiber type or a longer deployed urban loop with a fourth commercial O-band source; if the same fourth-degree A(λ) no longer predicts the location or depth of the minimum within the reported relative-error bounds, the claimed generality fails.","tokens_in":9909,"feed_emoji":"📡","tokens_out":943,"duration_ms":19904,"temperature":0.7,"pith_summary":"Quantum networks will have to share existing fiber with ordinary classical traffic. When classical light sits in the O-band (around 1310 nm) and quantum light sits in the C-band, spontaneous Raman scattering turns some classical photons into broadband noise that lands on the quantum channels. This paper measures that noise with commercial SFP transceivers and ordinary single-mode fiber spools up to 5 km, then fits a compact formula that predicts the noise from launch power, wavelength and length. The formula’s spectral shape is essentially the same for three different sources, so the quietest C-band channels (near 1535 nm) can be chosen without re-measuring every transmitter. The practical payoff is a simple engineering rule for placing entanglement or QKD channels on urban fiber that already carries classical traffic.","feed_headline":"Model finds quietest C-band slots for quantum light on shared fiber","feed_subtitle":"One polynomial predicts Raman noise from ordinary O-band SFPs so quantum channels can sit where the noise is lowest","key_machinery":"The compact count model C_SpRS(S_src, λ, L) with a source-independent fourth-degree polynomial A(λ) for the Raman spectral profile at fixed 1310 nm pump; it separates power scaling, wavelength shape and fiber-length attenuation so optimal C-band channels can be read off without new spectral scans.","core_discovery":"Spontaneous Raman scattering from O-band classical light into the C-band is captured by the source-scaled expression C_SpRS(S_src, λ, L) = S_src A(λ) (e^{-α_c L} – e^{-α_o L})/(α_o – α_c), where A(λ) is a single fourth-degree polynomial fixed once from experiment. The same A(λ) works across commercial SFPs and a narrow-linewidth laser, revealing a clear noise minimum near 1535 nm (ITU channel 44) that can be used to allocate quantum channels.","pith_inferences":["If the polynomial A(λ) proves stable across manufacturers, operators could publish a single look-up table of preferred quantum channels for any O-band classical plant.","The observed minimum may shift under temperature or aging of the fiber; a short field re-calibration protocol would turn the model into a live network-management tool.","Extending the same fitting procedure to multi-wavelength O-band loads would immediately give noise maps for fully loaded classical links."],"forward_implications":["Quantum-network operators can pre-select C-band DWDM channels near 1535 nm to minimize Raman noise from co-propagating 1310 nm classical traffic.","Noise budgets for entanglement distribution or QKD on shared urban fiber can be estimated from launch power and length alone, without repeated spectral characterization of every transmitter.","The same model supplies a quantitative figure of merit for deciding when O-band classical / C-band quantum co-propagation is tolerable versus when spatial or temporal isolation is required.","Filter-cascade design rules (number of WDMs) can be set by the same measurements that produced the model."],"fun_headline_variants":["Polynomial maps O-band Raman noise to quietest C-band quantum slots","SpRS model finds 1535 nm least noisy for quantum with O-band SFPs","One source-scaled formula picks lowest-noise C-band quantum channels","Raman profile from commercial SFPs guides C-band quantum allocation","Compact SpRS expression locates quiet C-band windows on shared fiber"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The single polynomial shape of the Raman spectrum measured on laboratory spools, together with two fixed attenuation numbers, stays accurate enough for other fibers and real urban links without being re-fit.","fun_headline_variants_meta":{"raw":{"variants":["Polynomial maps O-band Raman noise to quietest C-band quantum slots","SpRS model finds 1535 nm least noisy for quantum with O-band SFPs","One source-scaled formula picks lowest-noise C-band quantum channels","Raman profile from commercial SFPs guides C-band quantum allocation","Compact SpRS expression locates quiet C-band windows on shared fiber"]},"model":"grok-4.5","effort":"low","cost_usd":0.00553,"raw_usage":{"total_tokens":1524,"prompt_tokens":849,"num_sources_used":0,"completion_tokens":83,"cost_in_usd_ticks":55304000,"prompt_tokens_details":{"text_tokens":849,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":592,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":849,"tokens_out":83,"duration_ms":10710,"temperature":1.0,"reasoning_tokens":592,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-30T17:30:06.778280+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Measure SpRS counts versus C-band wavelength on a different fiber type or a longer deployed urban loop with a fourth commercial O-band source; if the same fourth-degree A(λ) no longer predicts the location or depth of the minimum within the reported relative-error bounds, the claimed generality fails.","supporting_citations":[],"review_version":1}