Pith. sign in

REVIEW 3 major objections 6 minor 53 references

Programmable Photonic Unitary Processor Enables Parametrized Differentiable Long-Haul Spatial Division Multiplexed Transmission

T0 review · 3 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 Programmable photonic processors in a multimode fiber link, tuned by gradient descent through a differentiable transmission model, cut the modal delay spread from ~25 ns to ~10 ns over 1300 km and shrink the receiver's MIMO equalizer load.

desk verdict A real first for programmable photonic DMD reduction, but the recirculating-loop setup leaves the multi-span generalization unproven. read the letter →

arxiv 2505.17381 v1 pith:OEFXSJC7 submitted 2025-05-23 physics.optics cs.LGphysics.app-ph

Mitsumasa Nakajima , Kohki Shibahara , Kohei Ikeda , Akira Kawai , Masaya Notomi , Yutaka Miyamoto , Toshikazu Hashimoto This is my paper · ORCID
classification physics.opticscs.LGphysics.app-ph
keywords parameterizedSDMtransmissiondifferentialmodedelayprogrammablephotonicunitaryprocessorClementsMZImeshdifferentiablemodelspatialdivisionmultiplexingMIMO-DSPcomplexitysilicaPLC
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

In long-haul transmission over multimode fiber, the spread of arrival times between spatial modes — differential mode delay — forces the receiver's digital MIMO equalizer to cover a long temporal window, and the computational cost grows with that window. This paper tries to establish that the optical channel itself can be made trainable: programmable photonic unitary processors placed at intermediate nodes, configured by gradient descent through a fully differentiable model of the fiber link, can suppress the delay spread in the optical domain before the receiver sees the signal. The authors built an 8×8 Clements-type silica planar-lightwave-circuit (PLC) mesh with 2.1 dB fiber-to-fiber loss, polarization independence, and fidelity above $R^2 = 0.96$ across the C-band; in a 1300-km three-mode-fiber experiment, the optimized matrix cut the measured delay spread from roughly 25 ns to roughly 10 ns at three wavelengths while the normalized generalized mutual information (NGMI) stayed above the 0.836 error-free threshold. If the claim holds, receiver DSP complexity, energy per bit, and post-processing latency in SDM networks can all shrink, and the same programmable mesh could later combine dispersion shaping with routing or mode-dependent-loss control.

What carries the argument

The argument is carried by two coupled objects. The first is the differentiable digital twin of the link: each span is $M_k(\omega) = S_k D_k(\omega) V_k^\dagger$, with $D_k$ holding the modal delays as frequency-dependent phases, and the group-delay operator $G(\omega) = -j \frac{d}{d\omega} \log M_{\mathrm{total}}(\omega)$ converts the full chain into a Hermitian matrix whose eigenvalues are the accumulated delays; the maximum eigenvalue spread (Eq. 9) is the training loss. The second is the Clements decomposition, which factorizes $U_k$ into 2×2 Mach–Zehnder blocks parameterized by physical phase shifts $\xi_1, \xi_2$, so that backpropagation updates realizable hardware settings while $U_k$ stays exactly unitary — and the same differentiable model doubles as the twin for machine-learning calibration of the fabricated mesh. The physical enabler is the 8×8 silica PLC Clements circuit: equalized path lengths remove the delay-line wavelength dependence typical of interferometric meshes, dual-arm thermo-optic MZIs suppress thermal crosstalk, and calibration infers residual coupler splitting ratios and external phase errors, lifting fidelity from $R^2 = 0.53$ to $0.99$ at 1550 nm and holding $R^2 > 0.96$ across the C-band — above the $R^2 \approx 0.95$ level the paper identifies as the degradation threshold.

What would settle it

Characterize the installed three-mode fiber's actual coupling and delay statistics from its measured transfer matrix and repeat the optimization on that measured model: if the resulting unitary no longer outperforms a random mode scrambler in shrinking the 95% equalizer window over 1300 km, the central claim fails. The paper's own simulations show that in the strong-coupling regime optimization gives no advantage over random scrambling, so checking which regime the real fiber sits in is the decisive observation.

Watch

Extended reading notes

Core claim

The paper's central claim is that differential mode delay — the arrival-time spread among spatial modes that sets the equalizer window length $L$ and hence the $O(N^2 L)$ cost of multiple-input multiple-output digital signal processing (MIMO-DSP) — can be suppressed in the optical domain by applying a wavelength-insensitive unitary matrix $U_k$ at each span, and that the right $U_k$ can be found by gradient-based optimization through a differentiable model rather than by heuristic or combinatorial search. The model writes each fiber span as $M_k(\omega) = S_k D_k(\omega) V_k^\dagger$, chains the spans with the trainable unitaries into $M_{\mathrm{total}}(\omega) = M_{K+1}(\omega) \prod_{k=1}^{K} U_k M_k(\omega)$, and reads the accumulated modal delays as the eigenvalues of the Hermitian group-delay operator $G(\omega) = -j \frac{d}{d\omega} \log M_{\mathrm{total}}(\omega)$, computed by singular value decomposition so the whole pipeline stays differentiable. Training then updates the MZI phase parameters of the Clements mesh by backpropagation, preserving unitarity exactly. The paper reports that the procedure converges in three-mode and ten-mode simulations, beats the random-mode-scrambling square-root-of-distance baseline in no-coupling and weak-coupling regimes, and — when the optimized matrix is transferred to the fabricated PLC mesh — reduces the experimentally measured 95% impulse-response window from about 25 ns to about 10 ns over 1300 km at 1541.349, 1550.057, and 1558.173 nm, with NGMI above the 0.836 threshold throughout.

Load-bearing premise

The load-bearing premise is that the digital twin of the fiber is faithful: the three-mode link really is weakly coupled between the LP01 and LP11 groups with negligible mode-dependent loss, and the delay spread the optimizer minimizes in simulation is the same quantity as the 95% pulse-energy window the real receiver's equalizer must cover — if either assumption fails, the optimized photonic setting will not cut the DSP load as predicted.

Editorial extensions

If this is right

  • Reducing the measured delay spread from ~25 ns to ~10 ns over 1300 km shortens the temporal window the receiver's MIMO-DSP must equalize, so digital post-processing load falls roughly in proportion to the window length.
  • Because DMD is a slow, statistical property of the link, the optimized unitary remains valid over long timescales and needs no fast tracking, keeping control overhead low in deployment.
  • Simulations show the suppression strengthens with denser processor deployment (shorter spans) and extends to ten-mode fibers with group-wise coupling, so the three-mode experiment is a scaled proof of a general mechanism.
  • The same differentiable model can target impairments beyond delay: the paper shows the optimized unitaries also shift mode-dependent loss and its NGMI penalty, opening the way to joint DMD/MDL optimization.
  • Idle ports of the Clements mesh can be repurposed for spatial-mode routing, so a single programmable node could eventually combine dispersion shaping, MDL control, and switching.

Reading between the lines

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

  • The fidelity budget the paper measures (roughly $R^2 > 0.95$ for acceptable DMD degradation) doubles as a portability criterion: any low-loss broadband mesh platform that meets it could host the same optimization, making the result a system-level recipe rather than a single-device demonstration.
  • Because the optimizer needs only a differentiable model, the same train-the-channel pattern should transfer to other slowly varying impairments — span-wise mode-dependent loss, gain ripple, or slow drift of the mesh itself — wherever a surrogate model can be written.
  • The paper observes many near-equivalent, initial-condition-dependent optima, which implies the cost landscape is flat along several directions; that degeneracy could be exploited as a free degree of freedom to choose unitaries that also improve routing or MDL without sacrificing delay suppression.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 6 minor

Summary. The paper proposes 'parameterized SDM transmission,' in which programmable photonic unitary processors placed at intermediate nodes of a spatial-division-multiplexed fiber link are configured, through a differentiable digital twin, to minimize accumulated differential mode delay and thereby reduce the MIMO-DSP equalization burden. The authors fabricate a silica-PLC 8×8 Clements-style unitary processor with 2.1 dB fiber-to-fiber insertion loss and R2>0.96 fidelity across the C-band, and they demonstrate a 1300-km three-mode recirculating-loop experiment in which an optimized unitary transformation reduces the measured 95% modal-dispersion window from about 25 ns to about 10 ns, with NGMI staying above the 0.836 error-free threshold. The paper also reports numerical optimization studies for three-mode and ten-mode fibers under different coupling scenarios and a series of supplementary analyses (calibration method, span-length dependency, comparison with simulated annealing and fixed permutations).

Significance. If the generalization claims hold, this is a notable step toward optical-domain MIMO preprocessing for long-haul SDM: it demonstrates a telecom-grade programmable unitary processor with state-of-the-art loss and bandwidth, and it provides a gradient-based framework for optimizing physical-layer transformations against a differentiable channel model. The deterministic-Θ sweep (Figs. 4) and the optimized-U experiment (Fig. 6) exhibit internally consistent agreement between the digital twin and measured DMD, and the device characterization in Fig. 3 is careful and convincing. The main significance is tempered by two validation weaknesses: the fiber model parameters are chosen to 'well explain' the same experimental results they are then used to validate, and the recirculating-loop experiment does not reproduce the statistically independent spans assumed in Eq. (4).

major comments (3)
  1. [Methods, Eq. (4) and experimental setup (Results, Fig. 4(a))] Eq. (4) models K statistically independent fiber spans, with each M_k(ω) drawn from independent random S_k and V_k, and the optimization in Fig. 2 averages over 100 such draws. In the recirculating loop, the same 51.2-km fiber and the same DEMUX/EDFA/U/MUX block are traversed K times, so the end-to-end operator is approximately M_{K+1}(U M)^K with a single fixed M. The Methods statement that the setup 'corresponds to testing the span-independent unitary matrix U' addresses only the span-independence of U, not the statistical independence of the spans. The agreement between a periodic-loop realization and an i.i.d.-span simulation average in Figs. 4–6 therefore does not establish that the same U would suppress DMD in a deployed link with physically distinct spans. Please provide a per-pass decorrelation measurement or a periodic-operator simulation demonstrating that the optimized U remains near-optimal, or explicitly limit the claims to periodic/recirculating-loop systems.
  2. [Methods, 'Numerical simulation and optimization' (three-mode case)] The key fiber-model parameters (LP11 modal delay of 55 ps/km, strong-intra-group/weak-inter-group coupling) are introduced with the statement that this choice 'well explains the experimental results,' but no independent characterization of the fiber is reported. Because the same experimental dataset is then used to validate the digital twin in Figs. 4–6, the agreement is partially circular and cannot serve as a predictive test of the model. Please provide an independent DMD/coupling measurement (for example OFDR or cutback characterization) or, if that is not available, a sensitivity analysis over the assumed parameters showing that the optimized unitary's DMD reduction is robust to the chosen values.
  3. [Methods, Eq. (10) and gradient-based optimization] The printed derivation of the gradient in Eq. (10) is incomplete: the group-delay operator is G(ω) = -j (dM/dω) M^{-1}, so dL/dU_k should involve the derivative of that full expression, including the M^{-1} factor and the -j; the finite-difference expression S'† (1/Δω)[dM(ω+Δω)/dU_k - dM(ω)/dU_k] V' lacks those terms. In practice the authors state that they use PyTorch autograd, which would differentiate through the full model and may be correct, but the equation as written does not support the method. Additionally, the cost function in Eq. (9) contains a maximum over modes, which is non-differentiable at ties; the manuscript does not discuss how autograd or the finite-difference approximation handles this. Please correct the derivation or state explicitly that autograd differentiates through G(ω) and describe the subgradient treatment, and confirm that the reported convergence is unaffected.
minor comments (6)
  1. [Figure 4 caption] The caption contains a garbled fragment: '(e) Implemented unitary matrices for different Θ setups. θ.' should be a proper sentence describing panel (b) or (e), and the panel labels in the caption do not match the figure panels.
  2. [Figure captions 4(e) and 6(a)] 'slid lines' should be 'solid lines' in both captions.
  3. [Figure 6 caption] The caption states 'we sent 2^XXX symbols,' which appears to be an unresolved placeholder; the main text says 33,360 symbols.
  4. [Supplementary Information S9] The section title contains a typo: 'Controlability of mode dependent loss uisng photonic unitary processor' should be 'Controllability ... using'.
  5. [Heading before Fig. 3] The heading 'SDM-WDM-PDM compatible programable photonic unitary processor' misspells 'programmable'.
  6. [Eqs. (1) and (2)] The matrix expressions for U_det and U_rot are misaligned in the rendered text; please ensure the row/column structure is typeset correctly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the optimized unitary is validated against an independent experimental DMD measurement, not fitted to it.

full rationale

The paper's central claim is that a unitary matrix optimized in a differentiable digital twin reduces the experimentally measured MIMO equalizer window from about 25 ns to about 10 ns over 1300 km. This is a genuine out-of-sample test: the optimizer minimizes the simulated group-delay spread of Eq. (9), and the resulting U is then physically implemented and evaluated against the measured 95% pulse-energy window, which is not used as a training target. The digital-twin parameters (55 ps/km modal delay and coupling conditions) are calibrated to the same fiber's baseline DMD-vs-Theta behavior, so the model is partly in-sample, but that calibration does not force the optimized U's performance; a wrong model or a poor optimizer could easily produce a unitary that fails to reduce the experimental window. The recirculating-loop setup tests a periodic operator rather than the independent spans of Eq. (4), which is a modeling-validity concern, but it is not circular because the experimental measurement is still independent of the optimization input. The self-citations (refs. 32, 41, 43, 45) are minor and not load-bearing: ref. 43 supplies a rotation-matrix parameterization, ref. 41 a chip-size reference, and refs. 32 and 45 background/support; none is invoked as a uniqueness theorem or as the justification for the central prediction. No equation is defined in terms of the claimed result, and no fitted parameter is renamed as a prediction. The paper is therefore self-contained against an external experimental benchmark for its headline DMD-reduction claim.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central DMD-reduction claim rests on a small set of fiber model parameters (the 55 ps/km delay and the qualitative coupling scenario) plus standard linear-algebra assumptions about unitary decomposition and group-delay statistics. No new physical entities are introduced. The 55 ps/km value is effectively calibrated to the baseline experimental behavior, which weakens the claim that the digital twin makes parameter-free predictions.

free parameters (3)
  • LP11 modal delay for three-mode fiber = 55 ps/km
    The paper states this value was set and that it 'well explains the experimental results', so it is effectively matched to the baseline DMD rather than independently measured in this work (Methods, numerical simulation paragraph).
  • Mode coupling strengths in QR-based unitary generation = qualitative (strong intra-group, weak inter-group)
    S13 uses hand-chosen block structures and small off-diagonal values to emulate the fiber coupling; the simulation results in Figs. 2 and 5 depend on this choice.
  • Ten-mode inter-group coupling scenarios = four qualitative cases (none, within degenerate modes, weak inter-group, strong all-mode)
    The 10-mode simulations in Fig. 2 explore these hand-set scenarios to claim scalability, but the exact coupling coefficients are not specified.
assumptions (5)
  • standard math Clements decomposition: any N x N unitary matrix can be realized by a mesh of MZIs with phase parameters (Eq. 5, ref 33).
    Used to parameterize U_k and to compute gradients through the circuit.
  • domain assumption Fiber span transfer matrix model M_k(omega) = S_k D_k(omega) V_k† (Eq. 6).
    Standard SDM model from Ho and Kahn (refs 34, 35); the optimization is only as good as this model.
  • domain assumption DMD is quasi-static: the statistical differential mode delay can be optimized with a fixed unitary per span without high-speed tracking.
    Stated in Methods; if true DMD fluctuates on fast timescales, the optimized settings would go stale.
  • domain assumption The singular values of the group-delay operator G(omega) correspond to the MIMO equalizer window used in the experiment (Eqs. 8-9 vs. the '95% region' definition in the Fig. 4 caption).
    The simulation optimizes Eq. 9 while the experiment measures the required equalizer length; the paper assumes these match.
  • ad hoc to paper QR-decomposition of a block-structured random matrix produces physically representative unitary coupling matrices S_k and V_k (S13).
    This generation method is introduced in this paper for computational efficiency; its fidelity to real fiber statistics is not independently validated.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Programmable Photonic Unitary Processor Enables Parametrized Differentiable Long-Haul Spatial Division Multiplexed Transmission." pith.science (2026). https://pith.science/paper/OEFXSJC7

@misc{pith2026250517381,
  author       = {Pith},
  title        = {Pith review of: Programmable Photonic Unitary Processor Enables Parametrized Differentiable Long-Haul Spatial Division Multiplexed Transmission},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OEFXSJC7}},
  note         = {Machine review of arXiv:2505.17381}
}
read the original abstract

The explosive growth of global data traffic demands scalable and energy-efficient optical communication systems. Spatial division multiplexing (SDM) using multicore or multimode fibers is a promising solution to overcome the capacity limit of single-mode fibers. However, long-haul SDM transmission faces significant challenges due to modal dispersion, which imposes heavy computational loads on digital signal processing (DSP) for signal equalization. Here, we propose parameterized SDM transmission, where programmable photonic unitary processors are installed at intermediate nodes. Instead of relying on conventional digital equalization only on the receiver side, our approach enables direct optimization of the SDM transmission channel itself by the programmable unitary processor, which reduces digital post-processing loads. We introduce a gradient-based optimization algorithm using a differentiable SDM transmission model to determine the optimal unitary transformation. As a key enabler, we first implemented telecom-grade programmable photonic unitary processor, achieving a low-loss (2.1 dB fiber-to-fiber), wideband (full C-band), polarization-independent, and high-fidelity (R2>96% across the C-band) operation. We experimentally demonstrate 1300-km transmission using a three-mode fiber, achieving strong agreement between simulation and experiment. The optimized photonic processor significantly reduces modal dispersion and post-processing complexity. Our results establish a scalable framework for integrating photonic computation into the optical layer, enabling more efficient, high-capacity optical networks.

Figures

Figures reproduced from arXiv: 2505.17381 by the authors.

Figure 1
Figure 1. Parameterized differentiable SDM transmission with photonic unitary processors. (a) Schematic explanation of parametrized SDM transmission. Statistical characteristics can be optimized by using trainable photonic unitary processors enabling direct optimization of the transmission matrix to reduce the pre/post MIMO processing load. (b) Optimization of photonic unitary conversion based on differential propagation mode… view at source ↗
Figure 2
Figure 2. Optimization of unitary matrix to reduce modal dispersion of MMF. Convergence process of modal dispersion over training epochs for (a) three and (b) ten- mode fibers under different coupling conditions of the MMF. Modal dispersion as a function of transmission distance for (c)–(f) three and (g)–(j) ten- mode fibers before (i.e., random matrixes drawn as blue lines) and after (red line) optimization of unitary matric… view at source ↗
Figure 1
Figure 1. Parameterized differentiable SDM transmission with photonic unitary processors. (a) Schematic explanation of parametrized SDM transmission. Statistical characteristics can be optimized by using trainable photonic unitary processors enabling direct optimization of the transmission matrix to reduce the pre/post MIMO processing load. (b) Optimization of photonic unitary conversion based on differential propagation mode… view at source ↗
Figures from the paper (1 more)
Figure 2
Figure 2. Figure 2: Optimization of unitary matrix to reduce modal dispersion of MMF. Convergence process of modal dispersion over training epochsfor (a) three and (b) ten- mode fibers under different coupling conditions of the MMF. Modal dispersion as a function of transmission distance …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 51 canonical work pages

  1. [1]

    New generation optical infrastructure technologies “EXA T initiative

    T. Morioka, “New generation optical infrastructure technologies “EXA T initiative” toward 2020 and beyond,” Proc. OECC2009, FT4, 2009

    work page 2020

  2. [2]

    Fiber-optic transmission and networking: the previous 20 and the next 20 years

    Winzer, P. J., Neilson, D. T. Neilson, and Chraplyvy , A . R. "Fiber-optic transmission and networking: the previous 20 and the next 20 years." Opti. Exp. 26, 24190-24239 (2018)

    work page 2018

  3. [3]

    Approaching the non-linear Shannon limit

    Ellis, A. D., Zhao, J. and Cotter, J. “Approaching the non-linear Shannon limit.” J. Light. Technol., 28,423–433 (2010)

    work page 2010

  4. [4]

    Few-mode fiber technology, deployments, and systems

    Sillard, P. et al. "Few-mode fiber technology, deployments, and systems." Proceedings of the IEEE 110, 1804-1820 (2022)

    work page 2022

  5. [5]

    Optical switching in future fiber- optic networks utilizing spectral and spatial degrees of freedom

    Marom, D. M., Miyamoto, Y ., Neilson, D. T., and Tomkos, I. “Optical switching in future fiber- optic networks utilizing spectral and spatial degrees of freedom.” Proceedings of the IEEE 110, 1835-1852 (2022)

    work page 2022

  6. [6]

    Space -division multiplexing for optical fiber communications

    Puttnam, B . J., Georg , R., and Ruben S. L. "Space -division multiplexing for optical fiber communications." Optica 8, 1186-1203 (2021)

    work page 2021

  7. [7]

    Mode -division multiplexing over 96 km of few -mode fiber using coherent 6× 6 MIMO processing

    Ryf, R. et al. "Mode -division multiplexing over 96 km of few -mode fiber using coherent 6× 6 MIMO processing." J. Light. Technol. 30, 521-531 (2012)

    work page 2012

  8. [8]

    22.9 Pb/s data -rate by extreme space -wavelength multiplexing

    Puttnam, B. J. et al. "22.9 Pb/s data -rate by extreme space -wavelength multiplexing." IET Conference Proceedings CP839. V ol. 2023. No. 34. Stevenage, UK: The Institution of Engineering and Technology, 2023

    work page 2023

Show all 53 references
  1. [9]

    DMD-unmanaged long-haul SDM transmission over 2500 -km 12-core× 3- mode MC -FMF and 6300 -km 3 -mode FMF employing intermodal interference canceling technique

    Shibahara, K. et al. "DMD-unmanaged long-haul SDM transmission over 2500 -km 12-core× 3- mode MC -FMF and 6300 -km 3 -mode FMF employing intermodal interference canceling technique." J. Light. Technol. 37, 138-147 (2019)

    2019

  2. [10]

    10.16-Peta-B/s dense SDM/WDM transmission over 6-mode 19-core fiber across the C+ L band

    Soma, D. et al. "10.16-Peta-B/s dense SDM/WDM transmission over 6-mode 19-core fiber across the C+ L band." J. Light. Technol. 36, 1362-1368 (2018)

    2018

  3. [11]

    10 -spatial-mode 1300-km transmission over 6 -LP graded index few -mode fiber with 36 -ns modal dispersion

    Shibahara, K., Megumi H ., and Yutaka M. "10 -spatial-mode 1300-km transmission over 6 -LP graded index few -mode fiber with 36 -ns modal dispersion." J Light . Technol. 42, 1257-1264 (2024)

    2024

  4. [12]

    Modal statistics in mode-division-multiplexed systems using mode scramblers

    Vijay, A., Krutko, O., Refaee, R., and Kahn, J. M. “Modal statistics in mode-division-multiplexed systems using mode scramblers.” J. Light. Technol. 43, 845 – 856 (2025)

    2025

  5. [13]

    Long-period fiber gratings for mode coupling in mode -division- multiplexing systems

    Askarov, D., and Kahn, J. M. “Long-period fiber gratings for mode coupling in mode -division- multiplexing systems.” J. Light. Technol. 33, 4032-4038 (2016)

    2016

  6. [14]

    Energy-efficient hybrid analog and digital precoding for mmWave MIMO systems with large antenna arrays

    Gao, X. et al. "Energy-efficient hybrid analog and digital precoding for mmWave MIMO systems with large antenna arrays." IEEE J. Sel. Areas Commun. 34, 998-1009 (2016)

    2016

  7. [15]

    Intelligent reflecting surface-assisted millimeter wave communications: Joint active and passive precoding design

    Wang, P., Fang, J., Y uan, X., Chen, Z., and Li, H. “Intelligent reflecting surface-assisted millimeter wave communications: Joint active and passive precoding design. ” IEEE Trans. V eh. Technol. 69, 14960-14973 (2020)

    2020

  8. [16]

    Recent developments of butler matrix from components design evolution to system integration for 5g beamforming applications: A survey

    Shallah, A. B. et al. "Recent developments of butler matrix from components design evolution to system integration for 5g beamforming applications: A survey." IEEE Access 10, 88434-88456 (2022)

    2022

  9. [17]

    Broadband physical layer cognitive radio with an integrated photonic processor for blind source separation

    Zhang, W., Tait, A., Huang, C. et al. “Broadband physical layer cognitive radio with an integrated photonic processor for blind source separation. ” Nat. Commun. 14, 1107 (2023). https://doi.org/10.1038/s41467-023-36814-4

    2023 doi

  10. [18]

    General-purpose programmable photonic processor for advanced radiofrequency applications

    Pérez-López, D., Gutierrez, A., Sánchez, D. et al. “General-purpose programmable photonic processor for advanced radiofrequency applications. ” Nat. Commun. 15, 1563 (2024). https://doi.org/10.1038/s41467-024-45888-7

    2024 doi

  11. [19]

    An integrated photonic-assisted phased array transmitter for direct fiber to mm -wave links

    Sanjari, P. and Aflatouni, F. “An integrated photonic-assisted phased array transmitter for direct fiber to mm -wave links.” Nat. Commun. 14, 1414 (2023). https://doi.org/10.1038/s41467 -023- 37103-w

    2023 doi

  12. [20]

    Ultrabroadband high -resolution silicon RF -photonic beamformer

    Pablo, M-C. et al. "Ultrabroadband high -resolution silicon RF -photonic beamformer." Nat. Commun. 15.1 (2024): 1433

    2024

  13. [21]

    Deep learning with coherent nanophotonic circuits

    Shen, Y . et al. “Deep learning with coherent nanophotonic circuits. ” Nat. Photon. 11, 441–446 (2017)

    2017

  14. [22]

    Photonics for artificial intelligence and neuromorphic computing

    Shastri, B. J. et al., “Photonics for artificial intelligence and neuromorphic computing .” Nat. Photon. 15, pp. 102–114 (2021)

    2021

  15. [23]

    Parallel convolution processing using an integrated photonic tensor core

    Feldmann, J. et al. “Parallel convolution processing using an integrated photonic tensor core. ” Nature 589, 52–58 (2021)

    2021

  16. [24]

    Linear programmable nanophotonic processors

    Harris, N. C. et al. “Linear programmable nanophotonic processors.” Optica 5, 1623–1631 (2018)

    2018

  17. [25]

    Large-scale photonic chiplet Taichi empowers 160 -TOPS/W artificial general intelligence

    Xu, Z. et al. " Large-scale photonic chiplet Taichi empowers 160 -TOPS/W artificial general intelligence." Science 384, 202-209 (2024)

    2024

  18. [26]

    Programmable photonic integrated meshes for modular generation of optical entanglement links

    Dong, M. et al. “Programmable photonic integrated meshes for modular generation of optical entanglement links.” npj Quantum Inf 9, 42 (2023)

    2023

  19. [27]

    16× 16 MZI -based photonic accelerator toward space and wavelength division multiplexing

    Ikeda, K. et al. "16× 16 MZI -based photonic accelerator toward space and wavelength division multiplexing." SPIE Emerging Topics in Artificial Intelligence (ETAI), (2024)

    2024

  20. [28]

    Programmable photonic unitary circuits for light computing

    Kim, K., Park, K., Park, H., Yu, S., Park, N., & Piao, X. “Programmable photonic unitary circuits for light computing.” Nanophotonics, (2025): https://doi.org/10.1515/nanoph-2024-0602

    2025 doi

  21. [29]

    All -optical mimo demultiplexing using silicon-photonic dual-polarization optical unitary processor

    Tanomura, R., et al. "All -optical mimo demultiplexing using silicon-photonic dual-polarization optical unitary processor." J. Light. Technol. 41.12 (2023): 3791-3796

    2023

  22. [30]

    SeyedinNavadeh, S. et al. Determining the optimal communication channels of arbitrary optical systems using integrated photonic processors. Nat. Photon. 18, 149–155 (2024)

    2024

  23. [31]

    High performance planar lightwave circuit devices for large capacity transmission

    Takahashi, H. “High performance planar lightwave circuit devices for large capacity transmission.” Opt. Exp. 19, B173 (2011)

    2011

  24. [32]

    Scalable reservoir computing on coherent linear photonic processor

    Nakajima, M., Tanaka, K. & Hashimoto, T. “Scalable reservoir computing on coherent linear photonic processor.” Commun. Phys. 4, 1–12 (2021)

    2021

  25. [33]

    R., Humphreys, P

    Clements, W. R., Humphreys, P. C., Metcalf, B. J., Kolthammer, W. S., and Walmsley, I. A. ”Optimal design for universal multiport interferometers.” Optica 3, 1460-1465 (2016)

    2016

  26. [34]

    Mode-dependent loss and gain: statistics and effect on mode-division multiplexing

    Ho, K. P ., and Kahn, J. M., "Mode-dependent loss and gain: statistics and effect on mode-division multiplexing." Opt. Exp. 19, 16612-16635 (2011)

    2011

  27. [35]

    Statistics of group delays in multimode fiber with strong mode coupling

    Ho, K. P. and Kahn, J. M. “Statistics of group delays in multimode fiber with strong mode coupling.” J. Light. Technol. 29, 3119-3128 (2012)

    2012

  28. [36]

    Enhancing Long-Haul 15-Mode Fiber Performance: Mode Permutation for Reduced Modal Dispersion

    Giammarco, D. S. et al. "Enhancing Long-Haul 15-Mode Fiber Performance: Mode Permutation for Reduced Modal Dispersion." J. Light. Technol. 43, 481-491 (2024)

    2024

  29. [37]

    Shape-optimized multi-mode interference for a wideband visible light coupler

    Sakamoto, J., Goh, T., Katayose, S., Kasahara, R. & Hashimoto, T. “Shape-optimized multi-mode interference for a wideband visible light coupler.” Opt. Commun. 443, 221 (2019)

    2019

  30. [38]

    Universal linear optics

    Carolan, J. et al. “Universal linear optics.” Science 349, 711 (2015)

    2015

  31. [39]

    High reliability of silica-based 1 × 8 optical splitter modules for outside plant

    Aratake, A. “High reliability of silica-based 1 × 8 optical splitter modules for outside plant. ” J. Light. Technol. 34, 27 (2016)

    2016

  32. [40]

    Optical frequency distribution using laser repeater stations with planar lightwave circuits

    Akatsuka, T. et al. "Optical frequency distribution using laser repeater stations with planar lightwave circuits." Opt. Exp. 28, 9186-9197 (2020)

    2020

  33. [41]

    High-fidelity WDM-compatible photonic processor for matrix-matrix multiplication

    Ikeda, K., Nakajima, M., Kita, S., Shinya, A., Notomi, M., and Hashimoto, T. “High-fidelity WDM-compatible photonic processor for matrix-matrix multiplication.” CLEO, JTh2A-87 (2024)

    2024

  34. [42]

    Non-duplicate polarization- diversity 8× 8 Si-wire PILOSS switch integrated with polarization splitter-rotators

    Tanizawa, K., Suzuki, K., Ikeda, K., Namiki, S., and Kawashima, H. “Non-duplicate polarization- diversity 8× 8 Si-wire PILOSS switch integrated with polarization splitter-rotators.” Opt. Exp. 25, 10885-10892 (2016)

    2016

  35. [43]

    Long-haul three-mode-multiplexed transmission employing on-chip mode permutation technique based on a programable photonic unitary processor

    Shibahara, K. et al., “Long-haul three-mode-multiplexed transmission employing on-chip mode permutation technique based on a programable photonic unitary processor”, OECC, 194 (2024)

    2024

  36. [44]

    Normalized generalized mutual information as a forward error correction threshold for probabilistically shaped QAM

    Cho, J., Schmalen, L., and Winzer, P. J. “Normalized generalized mutual information as a forward error correction threshold for probabilistically shaped QAM. ” 2017 European Conference on Optical Communication (ECOC). DOI: 10.1109/ECOC.2017.8345872

    2017

  37. [45]

    DMD-Unmanaged Long-Haul SDM Transmission Over 2500-km 12-core × 3-mode MC -FMF and 6300 -km 3 -mode FMF Employing Intermodal Interference Cancelling Technique

    Shibahara, K., Mizuno, T., Lee, D., Miyamoto, Y ., Ono, H., Nakajima, K.,Saitoh, S., Takenaga, and K., Saitoh, K. “DMD-Unmanaged Long-Haul SDM Transmission Over 2500-km 12-core × 3-mode MC -FMF and 6300 -km 3 -mode FMF Employing Intermodal Interference Cancelling Technique.” O...

    2018

  38. [46]

    Physical deep learning with biologically inspired training method: gradient -free approach for physical hardware

    Nakajima, M., Inoue, K., Tanaka, K., Kuniyoshi, Y ., Hashimoto, T., and Nakajima, K. “Physical deep learning with biologically inspired training method: gradient -free approach for physical hardware.” Nat. Commun. 13, 7847 (2022)

    2022

  39. [47]

    The forward-forward algorithm: Some preliminary investigations

    Hinton, G. "The forward-forward algorithm: Some preliminary investigations." arXiv:2212.13345 (2022)

    2022 arXiv

  40. [48]

    On a scalable path for multimode SDM transmission. J. Light. Technol

    Barbosa, F. A. and Ferreira, F. M. “On a scalable path for multimode SDM transmission. J. Light. Technol.” 42, 219-228 (2023)

    2023

  41. [49]

    Tunable efficient unitary neural networks (eunn) and their application to rnns

    Jing, L. et al. "Tunable efficient unitary neural networks (eunn) and their application to rnns." 34th International Conference on Machine Learning (ICML), 1733-1741 (2017)

    2017

  42. [50]

    PyTorch: An Imperative Style, High -Performance Deep Learning Library

    Paszke, A. et al. “PyTorch: An Imperative Style, High -Performance Deep Learning Library. ” Advances in Neural Information Processing Systems 32, 8024–8035 (2019). Acknowledgement This study. was partially supported by NICT, Japan, under commissioned research JP012368C01001. W...

    2019

  43. [51]

    intra-group coupling

    Initialize block-wise strong coupling structure: Each mode group (e.g., corresponding to a degenerate LP mode set) is assigned a block of strong internal coupling. Within each group, a random unitary matrix is generated using Haar sampling to reflect fully mixed modal interact...

  44. [52]

    The magnitude and sparsity of these off-diagonal entries control the strength and range of modal interactions, enabling fine- grained modeling of weakly coupled systems

    Introduce inter-group weak coupling: Weak crosstalk between adjacent mode groups is introduced by assigning small random complex values to the off -diagonal sub-blocks of matrix A. The magnitude and sparsity of these off-diagonal entries control the strength and range of modal...

  45. [53]

    10-spatial-mode 1300-km transmission over 6-LP graded index few-mode fiber with 36-ns modal dispersion

    Orthogonalization via QR decomposition: The resulting matrix A, which encodes both strong intra -group and weak inter-group modal coupling, is then orthogonalized via QR decomposition: A=QR, where Q is a unitary matrix and R is upper triangular. To ensure that the unitary matr...

    2024

Pith tools

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