pith. machine review for the scientific record. sign in

arxiv: 2605.04924 · v1 · submitted 2026-05-06 · 📡 eess.SP · cs.SY· eess.SY

Recognition: unknown

423.7 + 426.5 Tb/s GMI Bi-Directional HCF Transmission

Jiaqian Yang , Romulo Aparecido , Eric Sillekens , Ronit Sohanpal , Mindaugas Jarmolovi\v{c}ius , Zelin Gan , Yang Hong , Morteza Kamalian-Kopae
show 23 more authors
Abdallah Ali Shahab Bakhtiari Gorajoobi Ruben S. Lu\'is Daniele Orsuti Aleksandr Donodin Vitaly Mikhailov Jiawei Luo David J. DiGiovanni Nicolas Fontaine Lauren Dallachiesa Mikael Mazur Roland Ryf Haoshuo Chen David Neilson Ian D. Phillips Wladek Forysiak Sergei K. Turitsyn Hideaki Furukawa Jamie Gaudette David J. Richardson Benjamin J. Puttnam Robert I. Killey Polina Bayvel
Authors on Pith no claims yet

Pith reviewed 2026-05-08 15:46 UTC · model grok-4.3

classification 📡 eess.SP cs.SYeess.SY
keywords hollow core fiberbidirectional transmissionoptical fiber communicationshigh capacity transmissionOESCL bandgeneralized mutual informationsingle mode fiber
0
0 comments X

The pith

Hollow-core fiber supports same-wavelength bi-directional transmission over 60 km at aggregate rates of 423.7 and 426.5 Tb/s across 42.5 THz bandwidth.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper establishes that hollow-core fiber can carry ultra-high-capacity traffic in both directions simultaneously on identical wavelengths spanning the O, E, S, C, and L bands. It reports generalized mutual information values that match the best published unidirectional single-mode fiber results, for a total of roughly 850 Tb/s over a 60 km span. A reader would care because this suggests a practical route to doubling fiber capacity without extra strands or complex wavelength separation, addressing the growing demand for backbone bandwidth in data networks. The work focuses on demonstrating feasibility rather than long-term deployment details.

Core claim

We demonstrate OESCL-band same-wavelength bi-directional transmission over 60 km HCF with 42.5 THz bandwidth, achieving GMIs comparable with the highest unidirectional SMF data-rates in both directions, with an aggregate of 423.7 + 426.5 Tb/s.

What carries the argument

Same-wavelength bi-directional transmission over hollow-core fiber in the OESCL band, which uses the fiber's low nonlinearity and broad transparency window to minimize inter-direction interference while delivering high generalized mutual information rates.

If this is right

  • A single fiber pair could carry nearly twice the traffic of conventional unidirectional setups without wavelength reallocation.
  • High-capacity links could use fewer physical fibers while maintaining performance comparable to current single-mode standards.
  • Wideband OESCL operation becomes viable for bidirectional traffic, reducing the need for separate band management hardware.
  • Hollow-core fiber moves closer to practical use in scenarios where latency and nonlinearity matter most.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the approach scales to longer distances, network operators could halve the number of deployed fiber strands for equivalent capacity.
  • The result raises the question of whether similar bidirectional performance can be reached with real-time transceivers and standard amplifiers.
  • It suggests testing combinations with existing single-mode infrastructure at the endpoints to ease adoption.
  • Power consumption per bit might improve if hollow-core fibers avoid the nonlinear penalties that limit standard fiber at high powers.

Load-bearing premise

The measured generalized mutual information values will hold up as practical error-free rates once real forward error correction, bidirectional crosstalk, and any nonlinear effects across the full 60 km are taken into account.

What would settle it

A measurement of post-FEC bit error rates exceeding the GMI-predicted threshold in a live 60 km bidirectional HCF link at these aggregate rates would show the claimed capacities are not achievable.

Figures

Figures reproduced from arXiv: 2605.04924 by Abdallah Ali, Aleksandr Donodin, Benjamin J. Puttnam, Daniele Orsuti, David J. DiGiovanni, David J. Richardson, David Neilson, Eric Sillekens, Haoshuo Chen, Hideaki Furukawa, Ian D. Phillips, Jamie Gaudette, Jiaqian Yang, Jiawei Luo, Lauren Dallachiesa, Mikael Mazur, Mindaugas Jarmolovi\v{c}ius, Morteza Kamalian-Kopae, Nicolas Fontaine, Polina Bayvel, Robert I. Killey, Roland Ryf, Romulo Aparecido, Ronit Sohanpal, Ruben S. Lu\'is, Sergei K. Turitsyn, Shahab Bakhtiari Gorajoobi, Vitaly Mikhailov, Wladek Forysiak, Yang Hong, Zelin Gan.

Figure 1
Figure 1. Figure 1: (a) Experimental UWB SMF transmission [1–3, 12–16]. Filled/open markers: net/GMI throughput. signal processing (DSP) to combat gas-absorption alongside a record 42.5 THz total bandwidth, en￾abled by novel bismuth-doped fibre amplifiers (BD￾FAs) developed to suppport ∼100 nm O-band transmission, we achieved an aggregate Bi-Di GMI throughput of 423.7 Tb/s + 426.5 Tb/s and a de￾coded rate of 396.9 Tb/s + 399.… view at source ↗
Figure 2
Figure 2. Figure 2: Experimental diagram of Bi-Di HCF transmission. Numbers below the amplifiers: maximum powers (in dBm) of the high-power amplifiers. Inset: cross section of the HCF. test (CUT) was combined with two neighbouring channels and amplified to form a 3×32 GBaud slid￾ing test band with 1.33 GHz guard-band, based on a 33.33 GHz frequency grid. Dummy WDM sig￾nals were generated from spectrally-shaped ampli￾fied spon… view at source ↗
Figure 3
Figure 3. Figure 3: Fibre characterisations. (a) OTDR traces. (b) Fibre attenuation. (c) Measured RB coefficient. real-time oscilloscopes were used to acquire the signals in the O/E- and S/C/L-bands, respectively. Pilot-based DSP with 4 % overhead, adaptive rate decoding, and code rate puncturing (with granular￾ity of ∼ 0.01) were applied offline [20]. Transmission results The signal spectra before and after transmission, mea… view at source ↗
Figure 4
Figure 4. Figure 4: (a) Transmitted FW/BW and received FW spectra, and signal characteristics of bandwidth, number of channels, and FW/BW launch power. (b) O-band received (zoom). (c) L-band received (zoom) view at source ↗
Figure 5
Figure 5. Figure 5: (a) shows the SNR of the measured 1275 channels spanning 42.5 THz optical band￾width for both transmission directions, obtained by averaging the best three of five traces obtained per wavelength channel. Nine edge channels in the O- and E-bands were excluded due to poor transceiver performance or water absorption dis￾tortion. The reduced SNR observed in the centres of E- and L-bands corresponds to the wate… view at source ↗
read the original abstract

We demonstrate OESCL-band same-wavelength bi-directional transmission over 60 km HCF with 42.5 THz bandwidth, achieving GMIs comparable with the highest unidirectional SMF data-rates in both directions, with an aggregate of 423.7 + 426.5 Tb/s.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript demonstrates OESCL-band same-wavelength bi-directional transmission over 60 km of hollow-core fiber (HCF) with a 42.5 THz bandwidth. It achieves generalized mutual information (GMI) rates of 423.7 Tb/s and 426.5 Tb/s in the two directions, claiming these are comparable to the highest unidirectional single-mode fiber (SMF) data rates.

Significance. If the results hold, this is a significant experimental demonstration of ultra-high-capacity bi-directional transmission using HCF, which offers low latency and reduced nonlinearity compared to SMF. The aggregate rate exceeds 850 Tb/s while operating simultaneously in both directions over a wide bandwidth, and the choice of GMI as the figure of merit already incorporates measured impairments from the bidirectional configuration. This strengthens the case for HCF in future high-speed systems.

major comments (1)
  1. [Results] Results section: The reported GMI values are presented without error bars, statistical variability from repeated measurements, or explicit analysis of bidirectional crosstalk and nonlinear impairments over the full 60 km span; this information is load-bearing for validating the comparability to top unidirectional SMF records.
minor comments (2)
  1. [Abstract] Abstract: The exact frequency ranges or number of channels contributing to the stated 42.5 THz bandwidth should be specified for clarity.
  2. [Experimental Setup] The manuscript would benefit from a brief table summarizing the key experimental parameters (launch powers, amplifier types, DSP details) to aid reproducibility.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the positive assessment of our manuscript and the recommendation for minor revision. We address the single major comment below.

read point-by-point responses

  1. Referee: Results section: The reported GMI values are presented without error bars, statistical variability from repeated measurements, or explicit analysis of bidirectional crosstalk and nonlinear impairments over the full 60 km span; this information is load-bearing for validating the comparability to top unidirectional SMF records.

    Authors: The reported GMI values are obtained directly from the measured received constellations after propagation through the entire 60 km bi-directional HCF link; therefore the effects of bidirectional crosstalk and nonlinear impairments are already incorporated in the data. We agree that additional quantification would strengthen the presentation. In the revised manuscript we will add error bars derived from the variance across the measured sub-bands and include a concise analysis of the bidirectional crosstalk and nonlinear penalties relative to unidirectional operation. We did not perform multiple independent full-span repetitions, so formal statistical variability from repeated measurements cannot be supplied. revision: partial

Circularity Check

0 steps flagged

No circularity: experimental demonstration with external benchmarks

full rationale

The paper reports measured GMI values from a bi-directional OESCL-band transmission experiment over 60 km HCF, with aggregate rates of 423.7 + 426.5 Tb/s. No derivation chain exists; results follow directly from experimental measurements of constellations and impairments under simultaneous bidirectional operation. GMI is computed from the acquired data using standard formulas and is benchmarked against published unidirectional SMF rates from independent sources. No self-definitional steps, fitted inputs renamed as predictions, or load-bearing self-citations are present. The central claim remains independent and falsifiable against external experimental records.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

No free parameters or invented entities; the claim rests on standard assumptions about fiber propagation and GMI as a capacity proxy.

axioms (2)
  • domain assumption GMI provides a reliable upper bound on achievable information rate under the tested conditions
    Invoked implicitly when claiming comparability to highest unidirectional SMF rates.
  • domain assumption Bi-directional same-wavelength operation introduces no unaccounted crosstalk or nonlinear penalties over 60 km
    Required for the aggregate rate claim to hold without additional mitigation.

pith-pipeline@v0.9.0 · 5499 in / 1291 out tokens · 50347 ms · 2026-05-08T15:46:52.741745+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

22 extracted references · 14 canonical work pages

  1. [1]

    402 Tb/s GMI data-rate OESCLU- band transmission

    B. J. Puttnam et al. “402 Tb/s GMI data-rate OESCLU- band transmission”. In:Optical Fiber Communication Conference. 2024, Th4A.3.DOI: 10.1364/OFC.2024. Th4A.3

    work page doi:10.1364/ofc.2024 2024

  2. [2]

    450 Tb/s GMI, 42.4 THz, OESCL-Band Transmission Over a Field-Deployed Fiber

    R. S. Luís et al. “450 Tb/s GMI, 42.4 THz, OESCL-Band Transmission Over a Field-Deployed Fiber”. In:Optical Fiber Communication Conference. 2026, Th4B.5

    2026

  3. [3]

    430 Tb/S GMI Data Rate Over a Stan- dard G.654 Fiber Using Few-Mode O-Band and Single- Mode ESCL-Band Transmission

    R. S. Luís et al. “430 Tb/S GMI Data Rate Over a Stan- dard G.654 Fiber Using Few-Mode O-Band and Single- Mode ESCL-Band Transmission”. In:European Con- ference on Optical Communications. 2025, Th.03.02.3. DOI:10.1109/ECOC66593.2025.11263200

    work page doi:10.1109/ecoc66593.2025.11263200 2025

  4. [4]

    Broadband optical fibre with an attenuation lower than 0.1 decibel per kilometre

    Marco Petrovich et al. “Broadband optical fibre with an attenuation lower than 0.1 decibel per kilometre”. In: Nature Photonics19.11 (2025), pp. 1203–1208.DOI: 10.1038/s41566-025-01747-5

    work page doi:10.1038/s41566-025-01747-5 2025

  5. [5]

    Unrepeated HCF Transmission over spans up to 301.7 km

    A. Ali et al. “Unrepeated HCF Transmission over spans up to 301.7 km”. In:Optical Fiber Communications Con- ference. 2025, Th4A.3

    2025

  6. [7]

    Real-Time. Fully-Loaded C-band, Low- Latency, Long-Haul Transmission over Hollow-Core Fiber

    Y . Hong et al. “Real-Time. Fully-Loaded C-band, Low- Latency, Long-Haul Transmission over Hollow-Core Fiber”. In:European Conference on Optical Commu- nications. 2025, Th.03.02.2.DOI: 10.1109/ECOC66593. 2025.11262952

    work page doi:10.1109/ecoc66593 2025

  7. [8]

    Backscattering in antiresonant hollow-core fibers: over 40 dB lower than in standard optical fibers

    V. Michaud-Belleau et al. “Backscattering in antiresonant hollow-core fibers: over 40 dB lower than in standard optical fibers”. In:Optica8.2 (2021), pp. 216–219.DOI: 10.1364/OPTICA.403087

    work page doi:10.1364/optica.403087 2021

  8. [9]

    First Penalty-free Real-time Co- frequency Co-time Full-duplex Optical Fiber Transmis- sion with 202.1Tb/s Net Capacity Enabled by Hollow- core 5-element NANF

    Dawei Ge et al. “First Penalty-free Real-time Co- frequency Co-time Full-duplex Optical Fiber Transmis- sion with 202.1Tb/s Net Capacity Enabled by Hollow- core 5-element NANF”. In:Optical Fiber Communication Conference. 2024, M3J.2.DOI: 10.1364/OFC.2024.M3J. 2

    work page doi:10.1364/ofc.2024.m3j 2024

  9. [10]

    Field Trial of Real-time 128Tb/s Co- frequency Co-time Full-duplex Transmission over De- ployed 20km AR-HCFs in Urban Duct Network

    Dawei Ge et al. “Field Trial of Real-time 128Tb/s Co- frequency Co-time Full-duplex Transmission over De- ployed 20km AR-HCFs in Urban Duct Network”. In:Op- tical Fiber Communications Conference. 2025, W1C.4

    2025

  10. [11]

    502.6 Tbit/s S+C+L-Band Trans- mission in Anti-Resonant Hollow-Core Fiber

    Xumeng Liu et al. “502.6 Tbit/s S+C+L-Band Trans- mission in Anti-Resonant Hollow-Core Fiber”. In:Asia Communications and Photonics Conference. 2024, ACPIPOC-1014–14.DOI: 10 . 1109 / ACP / IPOC63121 . 2024.10809481

    work page arXiv 2024

  11. [12]

    273.6 Tbit/s real-time S+C+L- band same-wavelength bidirectional wavelength division multiplexing transmission in anti-resonant hollow-core fiber

    Xu Zhang et al. “273.6 Tbit/s real-time S+C+L- band same-wavelength bidirectional wavelength division multiplexing transmission in anti-resonant hollow-core fiber”. In:Optics Letters50.3 (2025), pp. 884–887.DOI: 10.1364/OL.538736

    work page doi:10.1364/ol.538736 2025

  12. [13]

    Tb/s/λ-Class Co-Frequency Co-Time Full-Duplex WDM Transmission Over 100-km AR-HCF in the S+C+L-band

    Siyuan Liu et al. “Tb/s/λ-Class Co-Frequency Co-Time Full-Duplex WDM Transmission Over 100-km AR-HCF in the S+C+L-band”. In:Journal of Lightwave T echnology 44.6 (2026), pp. 2076–2086.DOI: 10.1109/JLT.2026. 3651633

    work page doi:10.1109/jlt.2026 2026

  13. [14]

    Beyond 550 Tb/s S+C+L-band Bidirec- tional Transmission over 10.9-km Anti-Resonant Hollow- Core Fiber

    Xingfeng Li et al. “Beyond 550 Tb/s S+C+L-band Bidirec- tional Transmission over 10.9-km Anti-Resonant Hollow- Core Fiber”. In:Optical Fiber Communications Confer- ence. 2026, Th1J.6

    2026

  14. [15]

    339.1 Tb/s OESCLU-band trans- mission over 100 km SMF

    B. J. Puttnam et al. “339.1 Tb/s OESCLU-band trans- mission over 100 km SMF”. In:European Conference on Optical Communication. 2024, p. M.2.2

    2024

  15. [16]

    300 Tb/s O+S+C+L-Band Deployed Fibre Transmission

    Jiaqian Y ang et al. “300 Tb/s O+S+C+L-Band Deployed Fibre Transmission”. In:IEEE Photonics Conference. 2025, PD4.DOI:10.1109/IPC65510.2025.11282203

    work page doi:10.1109/ipc65510.2025.11282203 2025

  16. [17]

    Multiport E-band wavelength equal- izer

    N. K. Fontaine et al. “Multiport E-band wavelength equal- izer”. In:European Conference on Optical Communica- tions. 2023, pp. 1318–1321.DOI: 10.1049/icp.2023. 2535

    work page doi:10.1049/icp.2023 2023

  17. [18]

    O-band Bismuth Doped Fibre Ampli- fiers

    V. Mikhailov et al. “O-band Bismuth Doped Fibre Ampli- fiers”. In:European Conference on Optical Communica- tion. 2024, pp. 1408–1411

    2024

  18. [19]

    Turbulence-resilient oam-poisk with 21.92 db sensitivity gain in fsoc direct detection system

    Aleksandr Donodin et al. “High Power E-Band Bismuth- Doped Fiber Amplifier”. In:European Conference on Op- tical Communications. 2025, Tu.01.01.2.DOI: 10.1109/ ECOC66593.2025.11263326

    work page arXiv 2025

  19. [20]

    Transmission Over Field-Deployed Standard Single-Mode Fibre Using >100 nm S+C+L- Band

    Jiaqian Y ang et al. “Transmission Over Field-Deployed Standard Single-Mode Fibre Using >100 nm S+C+L- Band”. In:Journal of Lightwave T echnology43.13 (2025), pp. 6326–6334.DOI: 10 . 1109 / JLT . 2025 . 3546075

    2025

  20. [21]

    Gas Line Absorp- tion Mitigation in Hollow-Core Fibre using Spectral Pre- Equalisation

    Eric Sillekens and Ronit Sohanpal. “Gas Line Absorp- tion Mitigation in Hollow-Core Fibre using Spectral Pre- Equalisation”. In:Optical Fiber Communications Confer- ence. 2026, Th2A.50.URL: https://arxiv.org/abs/ 2602.09668

    work page arXiv 2026

  21. [22]

    High-Cardinality Geometrical Con- stellation Shaping for the Nonlinear Fibre Channel

    Eric Sillekens et al. “High-Cardinality Geometrical Con- stellation Shaping for the Nonlinear Fibre Channel”. In:Journal of Lightwave T echnology40.19 (2022), pp. 6374–6387.DOI:10.1109/JLT.2022.3197366

    work page doi:10.1109/jlt.2022.3197366 2022

  22. [23]

    The Case for a DNANF 1 Pb/s Trans-Atlantic Submarine Cable

    Pierluigi Poggiolini and Francesco Poletti. “The Case for a DNANF 1 Pb/s Trans-Atlantic Submarine Cable”. In:European Conference on Optical Communications. 2025, W.02.01.83.DOI: 10 . 1109 / ECOC66593 . 2025 . 11263039

    2025