Optical Network Digital Twin -- Practical Use Cases and Architecture

Andrea D'Amico; Daniel Kilper; Dmitrii Briantcev; Esther Le Rouzic; Gert Grammel; Giacomo Borraccini; Hideki Nishizawa; Juan Pedro Fernandez-Palacios; Kazuya Anazawa; Keigo Akahoshi

arxiv: 2511.06368 · v2 · submitted 2025-11-09 · 📡 eess.SY · cs.SY

Optical Network Digital Twin -- Practical Use Cases and Architecture

Hideki Nishizawa , Toru Mano , Kazuya Anazawa , Tatsuya Matsumura , Takeo Sasai , Masatoshi Namiki , Dmitrii Briantcev , Renato Ambrosone

show 12 more authors

Esther Le Rouzic Stefan Melin Oscar Gonzalez-de-Dios Juan Pedro Fernandez-Palacios Xiaocheng Zhang Keigo Akahoshi Gert Grammel Andrea D'Amico Giacomo Borraccini Marco Ruffini Daniel Kilper Vittorio Curri

This is my paper

Pith reviewed 2026-05-17 23:51 UTC · model grok-4.3

classification 📡 eess.SY cs.SY

keywords optical networksdigital twinquality of transmissionAI eranetwork managementGaussian noise modelinguse casesarchitecture

0 comments

The pith

An Optical Network Digital Twin architecture enables dynamic and Quality of Transmission aware operation for AI-era optical networks beyond conventional management.

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

The paper argues that rising AI-driven machine-to-machine traffic is changing what optical networks must do. It proposes a practical digital twin architecture that uses Gaussian noise modeling to simulate network behavior in real time and support quality-aware adjustments. This setup is meant to handle new operating modes such as operator optimization, user collaboration, and connections across multiple operators. The authors present concrete use cases and note the commercialization hurdles that still stand in the way. If the architecture works as described, optical networks could shift from static planning to continuous, predictive control.

Core claim

The paper proposes a practical Optical Network Digital Twin architecture that integrates recent advances in Gaussian noise modeling to enable dynamic, Quality of Transmission aware operation of optical networks in the AI era, going beyond conventional management approaches, with representative use cases in operator-driven optimization, user-operator collaboration, and multi-operator interconnection, while identifying key challenges for practical deployment.

What carries the argument

The Optical Network Digital Twin architecture, which models network elements and transmission performance to permit real-time, quality-of-transmission-aware adjustments and new operational workflows.

If this is right

Operator-driven optimization of routes and spectrum becomes feasible through continuous digital-twin simulation.
Users and operators can collaborate on service provisioning using shared twin predictions.
Interconnections between networks from different operators gain consistent quality monitoring.
Network management moves from periodic planning to ongoing dynamic control.

Where Pith is reading between the lines

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

Integration with existing control planes could shorten the time to first commercial use.
Similar modeling techniques might apply to other high-capacity transport layers facing AI traffic growth.
Validation would require testbeds that combine multiple vendors and real traffic patterns.
Success could reduce the need for over-provisioning and lower energy use per bit.

Load-bearing premise

The proposed architecture can overcome the significant commercialization barriers that currently exist for optical digital twins.

What would settle it

A pilot deployment in which the digital twin demonstrably improves network throughput, reduces outages, or enables new multi-operator services compared with existing management systems would support the claim; absence of such measurable gains in controlled field trials would challenge it.

Figures

Figures reproduced from arXiv: 2511.06368 by Andrea D'Amico, Daniel Kilper, Dmitrii Briantcev, Esther Le Rouzic, Gert Grammel, Giacomo Borraccini, Hideki Nishizawa, Juan Pedro Fernandez-Palacios, Kazuya Anazawa, Keigo Akahoshi, Marco Ruffini, Masatoshi Namiki, Oscar Gonzalez-de-Dios, Renato Ambrosone, Stefan Melin, Takeo Sasai, Tatsuya Matsumura, Toru Mano, Vittorio Curri, Xiaocheng Zhang.

**Figure 1.** Figure 1: Comparison of ONs and wireless networks the lowest layer of the OSI stack and supporting traffic from hundreds of thousands of users. Since these networks often span long-haul distances, link costs are substantial, setup may require several months, and the operational lifetime typically extends to around ten years. To honor the numerous service level agreements (SLAs) concluded with their customers, carrie… view at source ↗

**Figure 2.** Figure 2: Optical network digital twin architecture [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** Figure 3: QoT visualization [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

**Figure 4.** Figure 4: Changes in OLS types and operational methods [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗

**Figure 5.** Figure 5: Early use cases for optical network digital twins [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗

read the original abstract

With the widespread adoption of AI, machine-to-machine communications are rapidly increasing, reshaping the requirements for optical networks. Recent advances in Gaussian noise modeling for digital coherent transmission have raised expectations for digital-twin-based operation. However, unlike digital twins in wireless communication, which are already well established, significant barriers remain for commercialization in optical networks. This paper discusses the evolving requirements of optical networks in the AI era and proposes a practical Optical Network Digital Twin architecture enabling dynamic and Quality of Transmission aware operation beyond conventional management. Representative use cases, including operator-driven optimization, user-operator collaboration, and multi-operator interconnection, are presented, along with the architectural framework and key challenges toward practical deployment.

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.

Desk Editor's Note private letter to a colleague

High-level architecture sketch for optical digital twins that flags timely use cases but stays too conceptual to assess real practicality.

read the letter

The core point on this paper is that it maps out why digital twins could matter for optical networks under AI traffic growth and lists three use cases, yet it never shows how the pieces would actually run at the required speeds or with available data. It builds on recent Gaussian noise modeling work and contrasts the situation with more mature wireless digital twins, which is a fair framing. The use cases—operator-driven optimization, user-operator collaboration, and multi-operator interconnection—give a concrete sense of where the value might lie, and the discussion of commercialization barriers feels grounded in domain experience. That part is useful for anyone tracking network management evolution. The soft spots are exactly where the stress-test note points: the architecture is described at the block-diagram level with no equations for model ingestion or update rates, no latency budgets for the control loops, and no interface definitions that would let someone check against optical timing constraints. There is also no data, simulation, or even pseudocode to test whether the claimed dynamic QoT-aware operation is feasible. The citations follow standard prior work without obvious omissions, but the absence of any reproducible element means the practicality claim rests on untested assumptions. This is the kind of paper that could interest optical-network operators and researchers working on future control systems, especially those already thinking about digital-twin extensions. A reader looking for blueprints or measured results will come away empty. I would send it for peer review—the topic is relevant and the use-case framing is worth referee input, even if the manuscript needs substantial concrete additions to become actionable.

Referee Report

2 major / 2 minor

Summary. The paper discusses how AI-driven machine-to-machine traffic is reshaping optical network requirements and proposes a practical Optical Network Digital Twin (ONDT) architecture to enable dynamic, Quality-of-Transmission (QoT)-aware operation beyond conventional management. It outlines three representative use cases (operator-driven optimization, user-operator collaboration, and multi-operator interconnection), presents a high-level architectural framework with block diagrams, and identifies key challenges for deployment.

Significance. If the conceptual architecture can be realized with the required real-time performance, it would address an important gap by extending digital-twin concepts from wireless to optical networks and could guide the development of more adaptive, QoT-aware control systems in the AI era. The explicit enumeration of commercialization barriers and use cases provides a useful starting point for the community.

major comments (2)

[Architectural Framework] Architectural Framework section: the proposal states that Gaussian-noise models will support dynamic QoT-aware operation, yet no equations, data-flow diagrams, or latency budgets are supplied showing how model outputs are ingested, updated, and closed into control loops at line rate; without these the practicality claim cannot be evaluated.
[Use Cases] Multi-operator interconnection use case: the text asserts that data sharing can be reconciled with existing control-plane constraints, but provides neither interface definitions nor timing or security requirements that would allow an implementer to verify compatibility with current GMPLS or SDN control planes.

minor comments (2)

[Abstract] The abstract claims 'significant barriers remain for commercialization' but the main text does not enumerate or quantify those barriers, making it difficult for readers to assess how the proposed architecture specifically mitigates them.
[Figures] Figure captions for the block diagrams are brief; adding explicit labels for data paths, update frequencies, and control interfaces would improve clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their valuable feedback on our paper. We have prepared point-by-point responses to the major comments and made revisions to the manuscript to address the concerns raised regarding the architectural framework and use cases.

read point-by-point responses

Referee: Architectural Framework section: the proposal states that Gaussian-noise models will support dynamic QoT-aware operation, yet no equations, data-flow diagrams, or latency budgets are supplied showing how model outputs are ingested, updated, and closed into control loops at line rate; without these the practicality claim cannot be evaluated.

Authors: The manuscript presents a conceptual architecture for the Optical Network Digital Twin, building on recent advances in Gaussian noise modeling. We have revised the Architectural Framework section to include a high-level data-flow diagram illustrating how model outputs are integrated into the control loops. However, specific equations and precise latency budgets are not provided as they depend on particular hardware implementations and are identified as key challenges in the paper. This level of detail is beyond the scope of the current work, which focuses on the overall framework and use cases rather than a specific implementation. revision: partial
Referee: Multi-operator interconnection use case: the text asserts that data sharing can be reconciled with existing control-plane constraints, but provides neither interface definitions nor timing or security requirements that would allow an implementer to verify compatibility with current GMPLS or SDN control planes.

Authors: We agree that more specifics would be beneficial. In the revised manuscript, we have added a discussion on high-level interface considerations for multi-operator links, referencing compatibility with SDN-based control planes. Detailed interface definitions, timing, and security requirements are recognized as important for practical deployment and are listed among the commercialization barriers. Providing full specifications would require collaboration with standards bodies, which is outside the paper's scope but we highlight this as future work. revision: yes

Circularity Check

0 steps flagged

No circularity detected; paper is a high-level conceptual proposal without derivations or predictions.

full rationale

The manuscript presents a narrative discussion of optical network requirements in the AI era and outlines a proposed architecture with representative use cases, but supplies no equations, fitted parameters, Gaussian-noise model integrations, timing specifications, or first-principles derivations. All claims rest on domain assumptions and block-diagram descriptions rather than any chain that reduces by construction to its own inputs. No self-citation load-bearing steps, ansatz smuggling, or renaming of known results appear in the provided text. The work is therefore self-contained as an architecture sketch and receives the default non-circularity finding.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The proposal rests on the assumption that recent modeling advances raise expectations for digital-twin operation and that the listed barriers can be addressed through the proposed framework.

axioms (1)

domain assumption Recent advances in Gaussian noise modeling for digital coherent transmission enable digital-twin-based operation in optical networks
Stated in the abstract as raising expectations for digital-twin-based operation.

pith-pipeline@v0.9.0 · 5493 in / 1075 out tokens · 28360 ms · 2026-05-17T23:51:19.411404+00:00 · methodology

discussion (0)

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Leveraging Digital Twin Technologies: All-Photonics Networks-as-a-Service for Data Center Xchange in the Era of AI [Invited Tutorial]
eess.SY 2026-01 unverdicted novelty 4.0

Proposes DCX architecture with digital twins for all-photonics networks-as-a-service to create virtual large-scale data centers from distributed sites.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages · cited by 1 Pith paper

[1]

Aubrey, H

K. Aubrey, H. Wu, D. Mudigere, S. Anandaraj, and W. Gao. (2025, May) Turbocharge llm training across long-haul data center networks with nvidia nemo framework. Accessed: 2025-10-18. [Online]. Available: https://developer.nvidia.com/blog/turbocharge-llm-training- across-long-haul-data-center-networks-with-nvidia-nemo-framework/

work page 2025
[2]

(2024, Jul.) Remote media production for broadcast industry use case

IOWNGlobalForum. (2024, Jul.) Remote media production for broadcast industry use case. Accessed: 2025-10-18. [Online]. Available: https://iowngf.org/remote-media-production-for-broadcast- industry-use-case/

work page 2024
[3]

Gnpy model of the physical layer for open and disaggregated optical networking,

V . Curri, “Gnpy model of the physical layer for open and disaggregated optical networking,”Journal of optical communications and networking, vol. 14, no. 6, pp. C92–C104, 2022

work page 2022
[4]

Capacity and mutual information of wideband multipath fading channels,

I. E. Telatar and D. N. C. Tse, “Capacity and mutual information of wideband multipath fading channels,”IEEE Transactions on Information Theory, vol. 46, no. 4, pp. 1384–1400, July 2000

work page 2000
[5]

Open wireless digital twin: End-to-end 5g mobility emulation in o-ran framework,

T. Iye, M. Sakamoto, S. Takaya, E. Sato, Y . Susukida, K. Maruta, and J. Nakazato, “Open wireless digital twin: End-to-end 5g mobility emulation in o-ran framework,”arXiv preprint, Mar 2025

work page 2025
[6]

(2025) Chunghwa telecom and ericsson jointly advance 5g slicing automation for industrial digitalization

Ericsson press release. (2025) Chunghwa telecom and ericsson jointly advance 5g slicing automation for industrial digitalization. [Online]. Available: https://www.ericsson.com/en/press- releases/2/2025/2/chunghwa-telecom-and-ericsson-jointly-advance-5g- slicing-automation-for-industrial-digitalization

work page 2025
[7]

Applying digital twins to optical networks with cloud-native sdn controllers,

R. Valalta, L. Gifre, R. Casellas, R. Munoz, R. Martinez, A. Mozo, A. Pastor, D. Lopez, and J. P. Fernandez-Palacios, “Applying digital twins to optical networks with cloud-native sdn controllers,”IEEE communications magazine, vol. 61, no. 12, pp. 128–134, 2023

work page 2023
[8]

Experimental demonstration of partially disaggregated optical network control using the physical layer digital twin,

G. Borraccini, S. Straullu, A. Giorgetti, R. Ambrosone, E. Virgillito, A. D’Amico, R. D’Ingillo, F. Aquilino, A. Nespola, N. Samboet al., “Experimental demonstration of partially disaggregated optical network control using the physical layer digital twin,”IEEE Transactions on Network and Service Management, vol. 20, no. 3, pp. 2343–2355, 2023

work page 2023
[9]

Experimental validation of gnpy in a multi-vendor flex-grid flex-rate wdm optical transport scenario,

A. D’Amico, E. London, B. L. Guyader, F. Frank, E. L. Rouzic, E. Pincemin, N. Brochier, and V . Curri, “Experimental validation of gnpy in a multi-vendor flex-grid flex-rate wdm optical transport scenario,”J. Opt. Commun. Netw., vol. 14, no. 3, pp. 79–88, Mar 2022. [Online]. Available: https://opg.optica.org/jocn/abstract.cfm?URI=jocn- 14-3-79

work page 2022
[10]

Modeling transceiver ber-osnr characteristic for qot estimation in short-reach systems,

T. Mano, A. D’Amico, E. Virgillito, G. Borraccini, Y .-K. Huang, K. Anazawa, H. Nishizawa, T. Wang, K. Asahi, and V . Curri, “Modeling transceiver ber-osnr characteristic for qot estimation in short-reach systems,” in2023 International Conference on Optical Network Design and Modeling (ONDM). IEEE, 2023, pp. 1–3

work page 2023
[11]

Fast wdm provisioning with minimal probing: the first field experiments for dc exchanges,

H. Nishizawa, T. Mano, T. Ferreira de Lima, Y .-K. Huang, Z. Wang, W. Ishida, M. Kawashima, E. Ip, A. D’Amico, S. Okamotoet al., “Fast wdm provisioning with minimal probing: the first field experiments for dc exchanges,”Journal of Optical Communications and Networking, vol. 16, no. 2, pp. 233–242, 2024

work page 2024
[12]

Optical network tomography over live pro- duction network in multi-domain environment,

T. Sasai, K. Anazawa, D. Briantcev, D. Dass, S. Ali, G. Borraccini, Y . Kobatake, H. Ishihara, E. Kenny, G. Liu, T. Mano, M. Nakamura, H. Nishizawa, Y . Sone, E. Yamazaki, Y . Kisaka, K. Takasugi, K. Asahi, M. Ruffini, and D. Kilper, “Optical network tomography over live pro- duction network in multi-domain environment,” inEuropean Conference on Optical C...

work page 2025
[13]

Leveraging digi- tal twins for all-photonics networks-as-a-service: Enabling innovation and efficiency,

H. Nishizawa, K. Anazawa, T. Mano, and T. Sasai, “Leveraging digi- tal twins for all-photonics networks-as-a-service: Enabling innovation and efficiency,” inEuropean Conference on Optical Communication (ECOC), 2025. Hideki Nishizawais a Senior Research Engineer at NTT Network Innova- tion Laboratories, Japan. He received B.E. and M.E. degrees in Physics f...

work page 2025
[14]

Xiaocheng Zhangis a network architect at NTT DOCOMO BUSINESS, Inc

In September 2000, he joined Telefonica where he is currently leading the Metro and Core Transport Networks team with Telefonica Global Chief Technical Office. Xiaocheng Zhangis a network architect at NTT DOCOMO BUSINESS, Inc. He received a B.E. degree in engineering from Shanghai Jiao Tong University in 2014. He has been engaged in the development of sof...

work page 2000

[1] [1]

Aubrey, H

K. Aubrey, H. Wu, D. Mudigere, S. Anandaraj, and W. Gao. (2025, May) Turbocharge llm training across long-haul data center networks with nvidia nemo framework. Accessed: 2025-10-18. [Online]. Available: https://developer.nvidia.com/blog/turbocharge-llm-training- across-long-haul-data-center-networks-with-nvidia-nemo-framework/

work page 2025

[2] [2]

(2024, Jul.) Remote media production for broadcast industry use case

IOWNGlobalForum. (2024, Jul.) Remote media production for broadcast industry use case. Accessed: 2025-10-18. [Online]. Available: https://iowngf.org/remote-media-production-for-broadcast- industry-use-case/

work page 2024

[3] [3]

Gnpy model of the physical layer for open and disaggregated optical networking,

V . Curri, “Gnpy model of the physical layer for open and disaggregated optical networking,”Journal of optical communications and networking, vol. 14, no. 6, pp. C92–C104, 2022

work page 2022

[4] [4]

Capacity and mutual information of wideband multipath fading channels,

I. E. Telatar and D. N. C. Tse, “Capacity and mutual information of wideband multipath fading channels,”IEEE Transactions on Information Theory, vol. 46, no. 4, pp. 1384–1400, July 2000

work page 2000

[5] [5]

Open wireless digital twin: End-to-end 5g mobility emulation in o-ran framework,

T. Iye, M. Sakamoto, S. Takaya, E. Sato, Y . Susukida, K. Maruta, and J. Nakazato, “Open wireless digital twin: End-to-end 5g mobility emulation in o-ran framework,”arXiv preprint, Mar 2025

work page 2025

[6] [6]

(2025) Chunghwa telecom and ericsson jointly advance 5g slicing automation for industrial digitalization

Ericsson press release. (2025) Chunghwa telecom and ericsson jointly advance 5g slicing automation for industrial digitalization. [Online]. Available: https://www.ericsson.com/en/press- releases/2/2025/2/chunghwa-telecom-and-ericsson-jointly-advance-5g- slicing-automation-for-industrial-digitalization

work page 2025

[7] [7]

Applying digital twins to optical networks with cloud-native sdn controllers,

R. Valalta, L. Gifre, R. Casellas, R. Munoz, R. Martinez, A. Mozo, A. Pastor, D. Lopez, and J. P. Fernandez-Palacios, “Applying digital twins to optical networks with cloud-native sdn controllers,”IEEE communications magazine, vol. 61, no. 12, pp. 128–134, 2023

work page 2023

[8] [8]

Experimental demonstration of partially disaggregated optical network control using the physical layer digital twin,

G. Borraccini, S. Straullu, A. Giorgetti, R. Ambrosone, E. Virgillito, A. D’Amico, R. D’Ingillo, F. Aquilino, A. Nespola, N. Samboet al., “Experimental demonstration of partially disaggregated optical network control using the physical layer digital twin,”IEEE Transactions on Network and Service Management, vol. 20, no. 3, pp. 2343–2355, 2023

work page 2023

[9] [9]

Experimental validation of gnpy in a multi-vendor flex-grid flex-rate wdm optical transport scenario,

A. D’Amico, E. London, B. L. Guyader, F. Frank, E. L. Rouzic, E. Pincemin, N. Brochier, and V . Curri, “Experimental validation of gnpy in a multi-vendor flex-grid flex-rate wdm optical transport scenario,”J. Opt. Commun. Netw., vol. 14, no. 3, pp. 79–88, Mar 2022. [Online]. Available: https://opg.optica.org/jocn/abstract.cfm?URI=jocn- 14-3-79

work page 2022

[10] [10]

Modeling transceiver ber-osnr characteristic for qot estimation in short-reach systems,

T. Mano, A. D’Amico, E. Virgillito, G. Borraccini, Y .-K. Huang, K. Anazawa, H. Nishizawa, T. Wang, K. Asahi, and V . Curri, “Modeling transceiver ber-osnr characteristic for qot estimation in short-reach systems,” in2023 International Conference on Optical Network Design and Modeling (ONDM). IEEE, 2023, pp. 1–3

work page 2023

[11] [11]

Fast wdm provisioning with minimal probing: the first field experiments for dc exchanges,

H. Nishizawa, T. Mano, T. Ferreira de Lima, Y .-K. Huang, Z. Wang, W. Ishida, M. Kawashima, E. Ip, A. D’Amico, S. Okamotoet al., “Fast wdm provisioning with minimal probing: the first field experiments for dc exchanges,”Journal of Optical Communications and Networking, vol. 16, no. 2, pp. 233–242, 2024

work page 2024

[12] [12]

Optical network tomography over live pro- duction network in multi-domain environment,

T. Sasai, K. Anazawa, D. Briantcev, D. Dass, S. Ali, G. Borraccini, Y . Kobatake, H. Ishihara, E. Kenny, G. Liu, T. Mano, M. Nakamura, H. Nishizawa, Y . Sone, E. Yamazaki, Y . Kisaka, K. Takasugi, K. Asahi, M. Ruffini, and D. Kilper, “Optical network tomography over live pro- duction network in multi-domain environment,” inEuropean Conference on Optical C...

work page 2025

[13] [13]

Leveraging digi- tal twins for all-photonics networks-as-a-service: Enabling innovation and efficiency,

H. Nishizawa, K. Anazawa, T. Mano, and T. Sasai, “Leveraging digi- tal twins for all-photonics networks-as-a-service: Enabling innovation and efficiency,” inEuropean Conference on Optical Communication (ECOC), 2025. Hideki Nishizawais a Senior Research Engineer at NTT Network Innova- tion Laboratories, Japan. He received B.E. and M.E. degrees in Physics f...

work page 2025

[14] [14]

Xiaocheng Zhangis a network architect at NTT DOCOMO BUSINESS, Inc

In September 2000, he joined Telefonica where he is currently leading the Metro and Core Transport Networks team with Telefonica Global Chief Technical Office. Xiaocheng Zhangis a network architect at NTT DOCOMO BUSINESS, Inc. He received a B.E. degree in engineering from Shanghai Jiao Tong University in 2014. He has been engaged in the development of sof...

work page 2000