Pith. sign in

REVIEW 4 major objections 6 minor 55 references

The Dual Horizon: A Rendezvous of Computing and Communication Services at the Optical Layer in Optical Computing-Communication Integrated Network

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read By aggregating two lower-order optical signals into one higher-order signal at intermediate nodes, an optical network can serve computing requests with fewer wavelengths than optical-bypass routing on the COST239 topology.

desk verdict Interesting extension with a plausible quantitative result, but the missing MILP and the cost-free 16-QAM aggregation assumption keep the central claim conditional. read the letter →

arxiv 2505.18753 v1 pith:CJXKVZEW submitted 2025-05-24 cs.NI

Dao Thanh Hai , Isaac Woungang This is my paper · ORCID
classification cs.NI
keywords Opticalcommunication-computingintegratednetworkRoutingwavelengthandcomputingassignmentproblemOptical-computing-enabledIn-networkOptical-bypassOptical-layerintelligenceaggregationdesignplanning2.0
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proposes an optical computing-communication integrated network in which intermediate optical nodes do more than route lightpaths past each other; they can also perform computing operations on lightpath entities, specifically optically aggregating two QPSK lightpaths into a single 16-QAM lightpath. The aim is to serve computing requests, such as distributed machine-learning aggregation of partial results from two source data centers to a destination, directly at the optical layer. To quantify the benefit, the paper formulates the routing, wavelength and computing assignment (RWCA) problem, an extension of the traditional routing and wavelength assignment (RWA) problem, and solves it optimally on the realistic COST239 topology. The central claim is that the integrated approach achieves greater spectral efficiency, measured as fewer wavelengths, than the optical-bypass model, and the paper reports that the improvement holds for every destination node in the studied cases.

What carries the argument

The central object is the optical aggregation operation: an all-optical function, realized by coherent spectral superposition with electro-optic modulators, that combines two QPSK lightpaths (2 bits per symbol) into a single 16-QAM lightpath (4 bits per symbol) at an intermediate node. This operation converts what would be two wavelengths on shared links into one wavelength carrying the same combined information, which is the source of the spectral saving. It also introduces a new design dimension: choosing which node performs the computation. That choice, together with routing and wavelength assignment for the resulting lightpath segments, defines the routing, wavelength and computing assignment (RWCA) problem, which the paper models as a mixed integer linear program and contrasts with the traditional RWA problem.

What would settle it

Measure whether the 16-QAM lightpath produced by coherent spectral superposition of two QPSK lightpaths at an intermediate node requires extra spectral resources, such as an additional guard band, higher launch power, or a different wavelength, compared with the two separate QPSK lightpaths; if it does, re-solving the COST239 instances with that physical constraint will show whether the wavelength savings survive.

Watch

Extended reading notes

Core claim

The central claim is that integrating optical computing into the optical layer can reduce the minimum number of wavelengths needed to serve computing requests. In the studied scenario, each computing request has two source nodes and one destination: in the optical-bypass network the request is treated as two communication demands carried by two separate lightpaths to the destination, where aggregation happens electronically; in the integrated network, the two sources send lightpaths to a common intermediate node, which performs an optical aggregation operation that turns two QPSK lightpaths (2 bits per symbol each) into one 16-QAM lightpath (4 bits per symbol) carrying both partial results, and that single lightpath continues to the destination. The wavelength saving is demonstrated by optimal solutions on COST239: across all destination nodes the integrated network uses fewer wavelengths than optical-bypass, and in a detailed traffic instance it uses 2 wavelengths instead of 3. The paper also presents this as the first formulation of serving computing services directly at the optical layer in fiber-optic networks.

Load-bearing premise

The load-bearing premise is that any chosen intermediate node can aggregate two QPSK lightpaths into one 16-QAM lightpath at no extra spectral cost, with output capacity exactly equal to the combined input capacity, and with no physical-layer penalty.

Editorial extensions

If this is right

  • On the COST239 instances studied, the integrated network uses fewer wavelengths than optical-bypass for every destination node, with a representative case needing 2 wavelengths instead of 3.
  • A computing request with two sources and one destination is served by choosing a computing node and establishing three lightpaths, or two if the computing node is one of the sources, which is why the new problem is called routing, wavelength and computing assignment.
  • The spectral saving comes from replacing two QPSK lightpaths on shared links with one 16-QAM lightpath, so the number of wavelength-link units drops while the carried information is preserved.
  • Higher destination nodal degree reduces the minimum number of wavelengths in both designs because more incoming fibers allow more wavelength re-use.
  • Because RWA is NP-hard and RWCA has an order of magnitude more variables and constraints, RWCA inherits that complexity, so the spectral gain comes with a higher optimization cost.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Beyond the paper's two-to-one aggregation case, the same argument suggests that combining more than two lower-order lightpaths into one higher-order lightpath could give larger wavelength savings, but that scaling is not simulated.
  • The result is demonstrated for a many-to-one computing pattern (five source pairs, one destination); how the gain behaves under mixed communication and computing traffic or multiple simultaneous destinations remains an open question the paper does not address.
  • If physical-layer constraints force the aggregation to specific nodes or require extra guard bands, the wavelength saving may shrink; a sensitivity study varying the set of allowed computing nodes would test this.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes an 'optical computing-communication integrated network' in which two QPSK lightpaths arriving at an intermediate node are optically aggregated into a single 16-QAM lightpath, thereby providing in-network computation at the optical layer. It introduces a routing, wavelength, and computing assignment (RWCA) problem for serving computing requests in such networks, claims to formulate it as a mixed-integer linear program, and reports CPLEX-based results on the COST239 topology showing that the integrated design uses fewer wavelengths than the conventional optical-bypass design for all destination nodes. The motivating application is geo-distributed machine learning, where partial results from two source nodes are aggregated to a destination.

Significance. If the underlying optical aggregation operation is physically realizable at intermediate nodes with no additional spectral cost and no reach penalty, the paper would identify a new network design problem (RWCA) and a potentially useful architectural direction for in-network computing. The paper also gives a concrete traffic scenario and a comparative simulation setup, which are useful for framing the problem. However, the manuscript does not provide the claimed MILP formulation, does not justify the physical feasibility of the aggregation operation, and omits any physical-layer constraints (notably 16-QAM reach) that are essential for the reported wavelength savings to be credible. The paper's main quantitative claim therefore rests on unverified modeling assumptions rather than on a validated design methodology.

major comments (4)
  1. [Section 4] The paper states that 'the RWCA is formulated in the form of the mixed integer linear programming model' and that the model was solved by CPLEX, but no MILP is actually presented. Section 4 contains only a problem statement and Algorithm 1, which is a while-loop with no termination condition, no search procedure, and no substantive algorithmic content. Without the decision variables, objective function, and constraints (wavelength uniqueness, wavelength continuity, and one-computing-node selection are mentioned but never written mathematically), the CPLEX results in Section 5 cannot be independently checked or reproduced. This is a load-bearing omission because the paper's central comparison claims that 'all results for both designs are optimally collected.'
  2. [Section 3, Figure 1, and Section 5] The entire spectral-efficiency gain rests on the assumption that two QPSK lightpaths can be optically aggregated at any intermediate node into one 16-QAM lightpath with output capacity equal to the sum of the input capacities, at no additional spectral cost, no power penalty, and no reach penalty. The manuscript cites references [55,56] for coherent spectral superposition, but those works demonstrate modulation-format aggregation in a transmitter/electrical-signal context and do not establish that the operation can be performed on two arbitrary in-flight lightpaths at an intermediate network node. Moreover, the model includes no physical-layer constraint on the resulting 16-QAM lightpath: 16-QAM requires roughly 6.7 dB higher OSNR than QPSK at the same symbol rate and therefore has a shorter reach. In the COST239 topology, links can be several hundred kilometers long, and the 16-QAM lightpaths in Table 2 (e.g., 10-2-1) may exceed a realistic 16-QAM reach at 400 Gbps. If such lightpaths are physically infeasible, the claimed 'improvement is consistently realized across all destination nodes' could disappear or reverse.
  3. [Section 5, Figure 3] The traffic generation is underspecified. The text says that for each destination, 'from 10 remaining nodes forming 5 distinct computing requests (i.e., five pairs of sources (s1, s2))', but it does not state how the five pairs are chosen, whether the same five pairs are used for every destination node, or how many random instances were averaged. Since Figure 3 reports a single optimal wavelength count per destination, the claim that the improvement is 'consistently realized across all destination nodes' is only valid for the particular (unspecified) pair set. The authors should specify the exact traffic instances or report statistics over many randomly generated instances.
  4. [Section 4 and Section 5] The complexity statement that 'RWCA is one order of magnitude more complex in term of number of variables and constraints' than RWA is not backed by any variable/constraint counts or a formal complexity argument. While RWCA obviously extends RWA, the comparative complexity claim needs at least approximate counts from the actual formulation, which is currently absent.
minor comments (6)
  1. [Abstract] Typo: 'ligthpath' should be 'lightpath'.
  2. [Introduction, last paragraph] The paper refers to 'the Section 5' twice: once for Simulation Results and once for the summary/conclusion; the latter should be Section 6.
  3. [Algorithm 1] Algorithm 1 is not an algorithm in any standard sense; it is a loop over constraints with no search method, no initialization, and no termination condition. It should either be replaced with a description of the actual solution method (e.g., the CPLEX implementation details) or removed in favor of the explicit mathematical formulation.
  4. [Section 3, Figure 1 and Table 2] The notation '(2,3,1)' for a computing request is not explicitly defined in the text; it would be helpful to state that the triple is (s1, s2, d).
  5. [Section 2, contributions] The paper claims to be 'the first contribution tapping into the opportunity of providing computing services at the optical layer' but immediately lists several preceding works by the same authors ([50,51,52]) and another arXiv preprint [47] on integrated computation and communication. The novelty claim should be qualified to reflect the specific new contributions of this manuscript.
  6. [Section 3] The mechanism for optical aggregation at an intermediate node is described only in vague terms ('coherent spectral superposition', 'electro-optic modulators'). Since this is the key enabling technology, a more detailed physical explanation or a reference to a demonstration of in-network aggregation (as opposed to transmitter-side aggregation) is needed.

Circularity Check

1 steps flagged · score 6.0 of 10

Spectral-efficiency 'gain' is the paper's own definition of optical aggregation, so the COST239 comparison mostly restates the model's input.

  1. self definitional [Section 3 (illustrative example, p. 8); the same mechanism is optimized in Section 5, Fig. 3 and Tables 1-2]
    "The computing sense in this context should be interpreted as the addition of bits per symbol from two ordinary lightpaths (i.e., QPSK: 2 bits/symbol) to a new integrated one (i.e., 16QAM: 4 bits/symbol) as schematically illustrated in Fig. 1(c). ... The spectral cost for this approach would be one wavelength count (i.e., λ 1) and four wavelength-link units, suggesting a greater efficiency in harnessing the optical spectrum."

    The 'computing' operation is defined as exactly the replacement of two 2-bit/symbol QPSK lightpaths by one 4-bit/symbol 16-QAM lightpath. Since a 16-QAM channel at the same symbol rate carries exactly the combined bit rate of the two QPSK channels, the reduction from two wavelengths to one wavelength is a logical consequence of this definition, not an empirical finding. Section 5 then optimizes a model that encodes this aggregation as a free operation producing a single lightpath; the reported 'greater spectral efficiency' in Fig. 3 is therefore the same statement restated through the CPLEX solution rather than an independent result. The claim is built into the input model, not derived from independent physics or measurements.

full rationale

The paper's own definition of the novel computing operation is mathematically identical to the benefit it later reports. In Section 3, optical aggregation is defined as the addition of bits per symbol from two QPSK lightpaths (2 bits/symbol each) into one 16-QAM lightpath (4 bits/symbol). From that definition alone it follows that one aggregated lightpath replaces two lightpaths, and the paper immediately counts 'one wavelength count' versus the two needed in optical-bypass. Section 5's COST239 comparison (Fig. 3, Tables 1-2) is an exact optimization over this same model: the integrated formulation minimizes wavelengths under the constraint that aggregation produces a single lightpath carrying both source demands, so the savings are built into the input. This is a self-definitional reduction of the central 'greater spectral efficiency' claim. However, the paper does contain an independent, exact RWCA formulation and uses CPLEX to obtain optimums, so the combinatorial comparison is not fabricated; the circularity lies in presenting a definitional benefit as a demonstrated improvement. The self-citations [50]-[52] describe prior concept work but are not load-bearing for the numerical result, and the aggregation hardware is cited to external works [55,56]. Physical-layer reach of 16-QAM is a real correctness risk but not a circularity issue.

Assumptions & free parameters 0 free parameters · 5 assumptions · 2 invented entities

The central claim rests on the feasibility and zero-cost nature of optical aggregation, which is assumed rather than demonstrated at network scale. The RWCA model itself is not shown, so the entire simulation depends on an unverifiable formulation. The paper's own prior works [50,51,52] introduce the core concept, so the contribution is an extension of the authors' previous program.

assumptions (5)
  • domain assumption The two input lightpaths to an aggregation operation can be combined at any node into a single lightpath with modulation format whose capacity equals the sum of the two inputs (QPSK+QPSK -> 16QAM), at no additional spectral cost.
    Section 3 states that 'by exploiting the coherent spectral superposition, optical aggregation could be efficiently realized through electro-optic modulators' and the example uses QPSK to 16QAM conversion. This is a physical-capability assumption that is not derived or experimentally validated for arbitrary network nodes in this paper.
  • standard math Wavelength continuity and wavelength uniqueness constraints are enforced as in conventional RWA.
    Algorithm 1 in Section 4 specifies these constraints. This is standard for RWA.
  • domain assumption For each computing request, exactly one computing node is selected and it must not be the destination node.
    Section 4, constraint 'one computing node means the computing node serving a computing request must not be the destination node and for each computing demand, one node and only one node is selected'.
  • standard math The mixed integer linear programming model for RWCA, solved by CPLEX, is correct and complete.
    Section 5 states the implementation was done in Matlab and solved by CPLEX to optimality, but the model itself is not displayed in the paper, so the correctness rests on an unshown formulation.
  • domain assumption The traffic instance of five computing requests per destination on COST239 is representative enough to conclude general spectral efficiency gains.
    Section 5 describes the traffic generation only as 'one node is chosen as the destination d and from 10 remaining nodes forming 5 distinct computing requests'. The selection of the pairs is not detailed, so the result may depend on the chosen traffic instance.
invented entities (2)
  • In-network optical aggregation operation independent evidence
    purpose: Combine two incoming lightpaths at a common intermediate node into a single higher-order modulated lightpath carrying the combined data, enabling computing services at the lightpath scale.
    The cited works [55,56] demonstrate optical channel aggregation based on coherent spectral superposition, providing some external evidence. However, the paper assumes this can be done at any node with no spectral cost, which is an extrapolation beyond the cited demonstrations.
  • Optical computing-communication integrated network
    purpose: A new network architecture that provides both communication and computing services at the optical layer.
    This is the paper's proposed architecture; no experimental prototype or system-level validation is provided, so it is a conceptual entity without an independent falsifiable handle.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The Dual Horizon: A Rendezvous of Computing and Communication Services at the Optical Layer in Optical Computing-Communication Integrated Network." pith.science (2026). https://pith.science/paper/CJXKVZEW

@misc{pith2026250518753,
  author       = {Pith},
  title        = {Pith review of: The Dual Horizon: A Rendezvous of Computing and Communication Services at the Optical Layer in Optical Computing-Communication Integrated Network},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CJXKVZEW}},
  note         = {Machine review of arXiv:2505.18753}
}
read the original abstract

With the significant advancements in optical computing platforms recently capable of performing various primitive operations, a seamless integration of optical computing into very fabric of optical communication links is envisioned, paving the way for the advent of \textit{optical computing-communication integrated network}, which provides computing services at the ligthpath scale, alongside the traditional high-capacity communication ones. This necessitates a paradigm shift in optical node architecture, moving away from the conventional optical-bypass design that avoids lightpath interference crossing the same node, toward leveraging such interference for computation. Such new computing capability at the optical layer appears to be a good match with the growing needs of geo-distributed machine learning, where the training of large-scale models and datasets spans geographically diverse nodes, and intermediate results require further aggregation/computation to produce the desired outcomes for the destination node. To address this potential use case, an illustrative example is presented, which highlights the merit of providing in-network optical computing services in comparison with the traditional optical-bypass mode in the context of distributed learning scenarios taking place at two source nodes, and partial results are then optically aggregated to the destination. We then formulate the new \textit{routing, wavelength and computing assignment problem} arisen in serving computing requests, which could be considered as an extension of the traditional routing and wavelength assignment, that is used to accommodate the transmission requests. Simulation results performed on the realistic COST239 topology demonstrate the promising spectral efficiency gains achieved through the \textit{optical computing-communication integrated network} compared to the optical-bypass model.

Figures

Figures reproduced from arXiv: 2505.18753 by the authors.

Figure 1
Figure 1. Serving a computing request in the traditional optical-byp [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. COST239 network [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Performance Comparison communication integrated network features greater spectral efficiency compared to the con￾ventional optical-bypass one when it comes to supporting computing requests and such improvement is consistently realized across all destination nodes in our studied cases. It is worth noting that the spectral saving enabled by optical computing-communication inte￾grated network comes at the cost of solvi… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 33 canonical work pages

  1. [1]

    P. J. Winzer, D. T. Neilson, A. R. Chraplyvy, Fiber-optic transmis sion and networking: the previous 20 and the next 20 Opt. Express 26 (18) (2018) 24190–24239. doi:10.1364/OE.26.024190. URL http://www.opticsexpress.org/abstract.cfm?URI=oe-26 -18-24190

    doi:10.1364/oe.26.024190 2018

  2. [2]

    A. Lord, C. White, A. Iqbal, Future optical networks in a 10 year time frame, in: 2021 Optical Fiber Communications Conference and Exhibition (OFC), 2021, pp. 1–3

    work page 2021

  3. [4]

    P. J. Winzer, Capacity scaling through spatial parallelism: From su bsea cables to short-reach optical links, in: 2021 Optical Fiber Communications Conference and Exhibitio n (OFC), 2021, pp. 1–1

    work page 2021

  4. [5]

    P. J. Winzer, K. Nakajima, C. Antonelli, Scaling optical fiber capac ities [scanning the issue], Proceedings of the IEEE 110 (11) (2022) 1615–1618. doi:10.1109/JPROC.2022.3212229

    work page arXiv 2022

  5. [6]

    P. J. Winzer, The future of communications is massively parallel, J . Opt. Commun. Netw. 15 (10) (2023) 783–787. doi:10.1364/JOCN.496992. URL https://opg.optica.org/jocn/abstract.cfm?URI=jocn-1 5-10-783

    doi:10.1364/jocn.496992 2023

  6. [7]

    Shtaif, C

    M. Shtaif, C. Antonelli, A. Mecozzi, X. Chen, The information capa city of the fiber-optic channel: Bounds and prospects, in: 2024 Optical Fiber Communications Conf erence and Exhibition (OFC), 2024, pp. 1–3

    work page 2024

  7. [8]

    van den Hout, Ultra-wideband and space-division multiplexed op tical transmission systems, Phd thesis 1 (research tu/e / graduation tu/e), Electrical Engine ering, proefschrift

    M. van den Hout, Ultra-wideband and space-division multiplexed op tical transmission systems, Phd thesis 1 (research tu/e / graduation tu/e), Electrical Engine ering, proefschrift. (Feb. 2024). doi:10.6100/jnxx-6t19. 14

    work page doi:10.6100/jnxx-6t19 2024

  8. [9]

    A. Lord, S. J. Savory, M. Tornatore, A. Mitra, Flexible technolo gies to increase optical network capacity, Proceedings of the IEEE 110 (11) (2022) 1714–1724. doi:10.1109/JPROC.2022.3188337

    arXiv 2022

Show all 55 references
  1. [11]

    Saleh, J

    A. Saleh, J. M. Simmons, All-optical networking: Evolution, bene fits, challenges, and future vision, Proceedings of the IEEE 100 (5) (2012) 1105–1117. doi:10.1109/JPROC.2011.2182589

    2012

  2. [12]

    D. T. Hai, On routing, wavelength, network coding assignment, and protection configuration problem in optical-processing-enabled networks, IEEE Transactions on Ne twork and Service Management 20 (3) (2023) 2504–2514. doi:10.1109/TNSM.2023.3283880

    2023

  3. [13]

    Ghobadi, Emerging optical interconnects for ai systems, in : Optical Fiber Communication Confer- ence (OFC) 2022, Optica Publishing Group, 2022, p

    M. Ghobadi, Emerging optical interconnects for ai systems, in : Optical Fiber Communication Confer- ence (OFC) 2022, Optica Publishing Group, 2022, p. Th1G.1. doi:10.1364/OFC.2022.Th1G.1. URL https://opg.optica.org/abstract.cfm?URI=OFC-2022-Th 1G.1

    2022 doi

  4. [14]

    S. Yan, Z. Zhu, M. S. Glick, Z. Wu, K. Bergman, Accelerating dist ributed machine learning in disaggregated architectures in: Optical Fiber Communication Conference (OFC) 2022, Optica Pub lishing Group, 2022, p. Th1G.2. doi:10.1364/OFC.2022.Th1G.2. URL https://opg.optica.org/a...

    2022 doi

  5. [15]

    H. Gu, X. Yu, Y. Lu, H. Zou, S. Li, X-nest+: A high bandwidth and reconfigurable optical inter- connects for distributed machine learning and high-performance c omputing, in: 2023 Optical Fiber Communications Conference and Exhibition (OFC), 2023, pp. 1–3. doi:10.1364/OFC.2023.Th3D.1

    2023 doi

  6. [16]

    W. Li, G. Yuan, C. Wu, P. Huang, Z. Wang, A reconfigurable optic al network for distributed deep learning, in: 2023 Opto-Electronics and Communications Confe rence (OECC), 2023, pp. 1–3. doi:10.1109/OECC56963.2023.10209635

    2023

  7. [17]

    J. M. Simmons, G. N. Rouskas, Routing and Wavelength (Spectr um) Assignment, Springer Interna- tional Publishing, Cham, 2020, pp. 447–484. doi:10.1007/978-3-030-16250-4$\_$12 . URL https://doi.org/10.1007/978-3-030-16250-4$_$12

    2020 doi

  8. [18]

    Varma, J

    S. Varma, J. P. Jue, Regenerator placement and waveband ro uting in optical networks with impair- ment constraints, in: 2011 IEEE International Conference on Co mmunications (ICC), 2011, pp. 1–5. doi:10.1109/icc.2011.5963337

    2011

  9. [19]

    Martinelli, N

    F. Martinelli, N. Andriolli, P. Castoldi, I. Cerutti, All-optical regen erator placement in wson, in: 2014 International Conference on Optical Network Design and Modeling , 2014, pp. 299–304

    2014

  10. [20]

    Garrido, A

    C. Garrido, A. Leiva, A. Beghelli, A rmlsa algorithm with modulation f ormat conversion at intermediate nodes, in: 2017 19th International Conference on Transparent Optical Networks (ICTON), 2017, pp. 1–4. doi:10.1109/ICTON.2017.8025073

    2017

  11. [21]

    T. H. Dao, On optimal designs of transparent wdm networks wit h 1+1 protection leveraged by all-optical xor network coding schemes, Optical Fiber Technolog y 40 (2018) 93 – 100. doi:https://doi.org/10.1016/j.yofte.2017.11.009

    2018 doi

  12. [22]

    D. T. Hai, A bi-objective integer linear programming model for th e routing and network coding as- signment problem in wdm optical networks with dedicated protection , Computer Communications 133 (2019) 51 – 58. doi:https://doi.org/10.1016/j.comcom.2018.08.006

    2019 doi

  13. [23]

    D. T. Hai, On routing, spectrum and network coding assignment problem for transparent flex-grid optical networks with dedicated prote ction, Computer Communicationsdoi:https://doi.org/10.1016/j.comcom.2019.08.005

    2019 doi

  14. [24]

    D. T. Hai, An optimal design framework for 1+1 routing and netw ork coding assignment problem in wdm optical networks, IEEE Access 5 (2017) 22291–22298. doi:10.1109/ACCESS.2017.2761809

    2017

  15. [25]

    D. T. Hai, Leveraging the survivable all-optical wdm network des ign with network coding assignment, IEEE Communications Letters 21 (10) (2017) 2190–2193. doi:10.1109/LCOMM.2017.2720661

    2017

  16. [26]

    D. T. Hai, On solving the 1 + 1 routing, wavelength and network co ding assignment problem with a bi-objective integer Telecommunication Systems 71 (2) (2019) 155–165. doi:10.1007/s11235-018-0474-9 . URL https://doi.org/10.1007/s11235-018-0474-9

    2019 doi

  17. [27]

    D. T. Hai, Network coding for improving throughput in wdm optica l networks with dedicated protection, 15 Optical and Quantum Electronics 51 (387). doi:10.1007/s11082-019-2104-5 . URL https://doi.org/10.1007/s11082-019-2104-5

    doi

  18. [28]

    D. T. Hai, Photonic network coding and partial protection in opt ical-processing-enabled network: two for a tango, Optical and Quantum Electronics 54 (282). doi:10.1007/s11082-022-03628-5 . URL https://doi.org/10.1007/s11082-022-03628-5

    doi

  19. [29]

    D. T. Hai, Optical networking in future-land: from optical-bypa ss-enabled to optical-processing-enabled paradigm, Optical and Quantum Electronics 55 (864). doi:10.1007/s11082-023-05123-x . URL https://doi.org/10.1007/s11082-023-05123-x

    doi

  20. [30]

    D. T. Hai, L. H. Chau, N. T. Hung, A priority-based multiobjectiv e design for routing, spectrum, and network coding assignment problem in network-coding-enabled elas tic optical networks, IEEE Systems Journal 14 (2) (2020) 2358–2369. doi:10.1109/JSYST.2019.2938590

    2020

  21. [31]

    E. D. Manley, J. Deogun, L. Xu, D. R. Alexander, All-optical net work coding, Optical Communications and Networking, IEEE/OSA Journal of 2 (4) (2010) 175–191. doi:10.1364/JOCN.2.000175

    2010 doi

  22. [32]

    Savva, K

    G. Savva, K. Manousakis, G. Ellinas, A network coding optimizatio n approach for physical layer security in elastic optical networks, IEEE Transactions on Network and Ser vice Management 22 (2) (2025) 1145–

    2025

  23. [33]

    P. L. McMahon, The physics of optical computing, Nature Revie ws Physics 5. doi:10.1038/s42254-023-00645-5 . URL https://doi.org/10.1038/s42254-023-00645-5

    doi

  24. [34]

    Bogaerts, D

    W. Bogaerts, D. P´ erez, J. Capmany, D. A. B. Miller, J. Poon, D . Englund, F. Morichetti, A. Melloni, Programmable photonic circuits, Nature 586. doi:10.1038/s41586-020-2764-0 . URL https://doi.org/10.1038/s41586-020-2764-0

    doi

  25. [35]

    SeyedinNavadeh, M

    S. SeyedinNavadeh, M. Milanizadeh, F. Zanetto, G. Ferrari, M. Sampi- etro, M. Sorel, D. A. B. Miller, A. Melloni, F. Morichetti, Determining the optimal communication channels of arbitrary optica l systems using integrated photonic processors, Nature Photonics 18. doi:10.1038/...

    doi

  26. [36]

    Shekhar, W

    S. Shekhar, W. Bogaerts, L. Chrostowski, J. E. Bowers, M. H ochberg, R. Soref, B. J. Shastri, Roadmapping the next generation of silicon photonics, Nat ure Communications 15. doi:10.1038/s41467-024-44750-0 . URL https://doi.org/10.1038/s41467-024-44750-0

    doi

  27. [37]

    Pile, Optical computing and artificial intelligence, Nature Phot onics 18

    D. Pile, Optical computing and artificial intelligence, Nature Phot onics 18. doi:10.1038/s41566-024-01582-0 . URL https://doi.org/10.1038/s41566-024-01582-0

    doi

  28. [38]

    T. Fu, J. Zhang, R. Sun, Y. Huang, W. Xu, S. Yang, Z. Zhu, H. C hen, Optical neural networks: progress and challenges, Light Science & Applications 13 (1). doi:10.1038/s41377-024-01590-3 . URL https://www.nature.com/articles/s41377-024-01590-3# citeas

    doi

  29. [39]

    H. Wang, J. Hu, A. Morandi, A. Nardi, F. Xia, X. Li, R. Savo, Q. Liu , R. Grange, S. Gigan, Photonics breakthroughs 2024: Nonlinear photonic computing at scale, IEEE Photonics Journal 17 (2) (2025) 1–4. doi:10.1109/JPHOT.2025.3547948

    2025

  30. [40]

    T. Zhou, F. Scalzo, B. Jalali, Nonlinear schr¨ odinger kernel for hardware acceleration of machine learning, Journal of Lightwave Technology 40 (5) (2022) 1308–1319. doi:10.1109/JLT.2022.3146131

    2022

  31. [41]

    Zhong, M

    Z. Zhong, M. Yang, J. Lang, C. Williams, L. Kronman, A. Sludds, H . Esfahanizadeh, D. Englund, M. Ghobadi, Lightning: A reconfigurable photonic-electronic smart nic for fast and energy-efficient inference, in: Proceedings of the ACM SIGCOMM 2023 Conference, ACM SIGCOM M ’23, Asso...

    2023

  32. [42]

    M. Yang, Z. Zhong, M. Ghobadi, On-fiber photonic computing, in : Proceedings of the 22nd ACM Workshop on Hot Topics in Networks, HotNets ’23, Association for C omputing Machinery, New York, NY, USA, 2023, p. 263–271. doi:10.1145/3626111.3628177. 16 URL https://doi.org/10.1145/3...

    2023

  33. [43]

    Sludds, S

    A. Sludds, S. Bandyopadhyay, Z. Chen, Z. Zhong, J. Cochran e, L. Bernstein, D. Bunandar, P. B. Dixon, S. A. Hamilton, M. Streshinsky, A. Novack, T. Baehr-Jones , M. Hochberg, M. Ghobadi, R. Hamerly, D. Englund, Delocalized photonic deep learning on the inte rnet’s edge, Scienc...

    2022 doi

  34. [44]

    L. Liu, L. Song, X. Chen, H. Yu, G. Sun, Accelerating model syn chronization for distributed machine learning in an optical wide area network, Journal of Optical Commun ications and Networking 14 (10) (2022) 852–865. doi:10.1364/JOCN.462286

    2022 doi

  35. [45]

    L. Liu, H. Yu, G. Sun, L. Luo, Q. Jin, S. Luo, Job scheduling for d istributed machine learning in optical wan, Future Generation Computer Systems 112 (2020) 549–560. doi:https://doi.org/10.1016/j.future.2020.06.007. URL https://www.sciencedirect.com/science/article/pii/S01677...

    2020 doi

  36. [46]

    M. Lian, Y. Zhao, X. Li, W. Liu, Y. Li, M. Tornatore, J. Zhang, Re source allocation in flexible- bandwidth fine-grained optical transport networks for geo-dist ributed machine learning, IEEE Internet of Things Journal (2025) 1–1 doi:10.1109/JIOT.2025.3558933

    2025

  37. [47]

    Zhang, L

    J. Zhang, L. Zhang, X. Pang, O. Ozolins, Q. Zhang, X. Yu, Integrated computation and communication with fiber-optic trans missions (2025). arXiv:2503.02165. URL https://arxiv.org/abs/2503.02165

    2025 arXiv

  38. [48]

    Zhang, D

    W. Zhang, D. Yang, C. Zhang, Q. Ye, H. Zhang, X. Shen, (com) 2net: A novel communi- cation and computation integrated network architecture, IEEE N etwork 38 (2) (2024) 35–44. doi:10.1109/MNET.2024.3355922

    2024

  39. [49]

    D. Wen, Y. Zhou, X. Li, Y. Shi, K. Huang, K. B. Letaief, A survey on integrated sens- ing, communication, and computation, IEEE Communications Survey s & Tutorials (2024) 1– 1doi:10.1109/COMST.2024.3521498

    2024

  40. [50]

    D. T. Hai, M. Nguyen, I. Woungang, Optical-computing-enabled network: A new dawn for optical-layer intelligence?, in: Proceedings of the 8th Asia-Pacific Workshop on Networking, AP Net ’24, Association for Comput- ing Machinery, New York, NY, USA, 2024, p. 215–216. doi:10.1145/...

    2024

  41. [51]

    D. T. Hai, What comes after optical-bypass network? a study o n optical-computing-enabled network, Optical Fiber Technology 84 (2024) 103730. doi:https://doi.org/10.1016/j.yofte.2024.103730. URL https://www.sciencedirect.com/science/article/pii/S1068520024000750

    2024

  42. [52]

    D. T. Hai, I. Woungang, Next-generation optical networks to sustain connectivity of the future: All roads lead to optical-compu ting-enabled Internet of Things 29 (2025) 101445. doi:https://doi.org/10.1016/j.iot.2024.101445. URL https://www.sciencedirect.com/science/article/pi...

    2025

  43. [53]

    W. Li, G. Yuan, Z. Wang, G. Tan, P. Zhang, G. N. Rouskas, Fast and scalable all-optical network architecture for distributed deep learning, J. Opt. Commun. Netw. 16 (3) (2024) 342–357. doi:10.1364/JOCN.511696. URL https://opg.optica.org/jocn/abstract.cfm?URI=jocn-1 6-3-342

    2024 doi

  44. [54]

    H. Zhu, W. Jiang, Q. Hong, Z. Guo, When in-network computing m eets distributed machine learning, IEEE Network 38 (5) (2024) 238–246. doi:10.1109/MNET.2024.3368138

    2024

  45. [55]

    Misra, S

    A. Misra, S. Preußler, K. Singh, J. Meier, T. Schneider, Optical channel aggregation based on modulation format convers ion by coherent spectral superposition with electro-optic APL Photonics 8 (8) (2023) 086112. arXiv:https://pubs.aip.org/aip/app/article-pdf/doi/ 10.1063/5.01...

    2023 doi

  46. [56]

    Misra, S

    A. Misra, S. Preussler, K. Singh, J. Meier, T. Schneider, Modula tion format aggregation of nyquist channels by spectral superposition with electro-optic modulators , in: 2022 Optical Fiber Communica- tions Conference and Exhibition (OFC), 2022, pp. 1–3. 17

    2022

  47. [1159]

    doi:10.1109/TNSM.2024.3498108

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.