Pith. sign in

REVIEW 3 major objections 3 minor 1 cited by

High Contrast Nulling in Photonic Meshes Through Architectural Redundancy

T0 review · 3 major / 3 minor · reviewed 2026-08-08 · deepseek-v4-flash

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

Pith's one-line read A triangular mesh of six double Mach-Zehnder interferometers suppresses light at one output by more than 80 decibels.

desk verdict A useful architecture that clearly improves nulling to a measured 72.5 dB, with a floor-limited >80 dB headline that needs better metrology before it convinces. read the letter →

arxiv 2502.07997 v2 pith:HQSGQ67T submitted 2025-02-11 physics.optics

Carson Valdez , Zhanghao Sun , Anne R. Kroo , David A. B. Miller , Olav Solgaard This is my paper · ORCID
classification physics.optics
keywords extinctionratioDoubleMach-Zehnderinterferometerphotonicmeshsiliconphotonicsself-configurationhigh-contrastnullingintegratedcoronagraphy
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 argues that a photonic circuit can suppress light at one output port far more deeply than a single interferometer can, and demonstrates it with a silicon chip. The architecture is a three-by-three triangular mesh of Double Mach-Zehnder Interferometers, configured one node at a time so that successive elements filter the leakage left by earlier ones. The authors report a single-channel extinction ratio above 80 dB, against 42.0 dB for a standard Mach-Zehnder interferometer and 62.4 dB for a standalone DMZI on the same chip. That matters for high-contrast applications such as space-based coronagraphy, which need rejection ratios of order $10^{-10}$ and need them in a compact, integrable form.

What carries the argument

The load-bearing object is the Double Mach-Zehnder Interferometer, a standard MZI in which each fixed coupler is replaced by an MZI whose internal phase shifter tunes the split ratio. The paper models a real coupler's split-ratio error as $\sigma$ and shows that a plain MZI's null is capped by that error, while a DMZI's two tunable splitters can be driven to cancel it and act as an effective 50:50 splitter. The mesh cascades three DMZIs along the light path so each stage rejects residual leakage from the previous one, and off-path DMZIs compensate parasitic paths from back reflections. A sequential algorithm configures devices one at a time: minimize one output, adjust the two tunable splitters with the same polarity, minimize the other output, adjust them with opposite polarity, then iterate until convergence.

What would settle it

Re-measure the optimized null with a calibrated detector whose noise floor is characterized below 1 pW, and check whether the null-port power scales with input power when the input is reduced; if the detected minimum remains pinned at the old detector's floor rather than falling with input, the true extinction has not been shown to exceed 80 dB.

Watch

Extended reading notes

Core claim

The central claim, stated on the authors' terms, is that architectural redundancy, not improved fabrication, is what enables high extinction. A three-stage cascade of DMZIs in a triangular mesh reaches a single-channel extinction ratio above 80 dB, two orders of magnitude better than a single DMZI at 62.4 dB and roughly 40 dB better than a standard MZI at 42.0 dB. The paper also reports that DMZIs lying off the nominal light path improve the null, which it attributes to back reflections at the coupling interfaces: those reflected fields take unintended routes through the mesh and can be cancelled by the extra tunable devices.

Load-bearing premise

The 80 dB result rests on the assumption that the photodiode's nominal 1 pW minimum detectable power is a calibrated upper bound on the actual optical power at the null port, and the paper does not report dark-current subtraction, noise-floor calibration, or repeated measurements at that floor.

Editorial extensions

If this is right

  • Extinction ratios beyond 80 dB are reachable in a silicon photonic mesh without exotic fabrication, using the same foundry process as a standard MZI plus programmable phase shifters.
  • The sequential tuning order means adding more stages to a triangular mesh is a straightforward extension: each new diagonal element is configured and frozen in turn, so deeper nulls scale with mesh size.
  • The measured improvement from off-path elements implies that parasitic reflections, normally treated as a fixed loss or noise source, can be actively nulled by redundant mesh elements.
  • A single-stage extinction of about 80 dB corresponds to a contrast near $10^{-8}$, narrowing the gap to the $10^{-10}$ contrast needed for direct exoplanet spectroscopy and suggesting that cascaded meshes could close it.
  • The same triangular mesh can serve as a building block for other high-contrast integrated photonic systems beyond coronagraphy, such as interferometric sensors that need deep nulls.

Reading between the lines

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

  • If the true optical null is as deep as measured, cascading more than three nulling stages should push extinction further, with the practical ceiling set by phase-shifter bit depth and detector dark noise rather than by split-ratio error.
  • The off-path node effect points to a direct test: fabricate the same mesh with couplers engineered to have different back-reflection levels and check whether the benefit from off-path DMZIs scales with reflection strength.
  • Because the headline number sits at the detector's nominal 1 pW floor, a fair verification would repeat the null measurement with a calibrated low-noise detector or use a second nulling mesh as a null-depth analyzer; until then, the 80 dB claim is best read as a lower bound set by the measurement, not necessarily the optical limit.
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

3 major / 3 minor

Summary. The paper reports a silicon-photonics experiment in which a 3×3 triangular mesh of six Double Mach-Zehnder Interferometers (DMZIs) is configured sequentially to null one output port. Standalone test devices give extinction ratios of 42.0 dB (MZI) and 62.4 dB (DMZI), consistent with prior work. In the mesh, sequential optimization of devices along the intended light path deepens the null from 63.3 dB to 72.5 dB, and optimizing an off-light-path device reduces the detected power below the stated 1 pW nominal detection limit of the photodiode, corresponding to a nominal 81.8 dB extinction ratio. The authors attribute the off-light-path improvement to back reflections from coupling interfaces and claim a single-channel extinction ratio exceeding 80 dB, with applications to high-contrast sensing and coronagraphy.

Significance. If validated, the architecture is significant: it extends the well-known DMZI nulling approach by cascading multiple DMZIs in a triangular mesh and using redundant off-path devices, with a convergent sequential configuration algorithm. The measured sequence 63.3 dB → 72.5 dB → detector floor is internally consistent and the standalone MZI/DMZI values agree with the literature. The paper also honestly flags, in the Figure 4 caption, that apparent rejection beyond 81.8 dB is not reliable. However, the central quantitative claim of exceeding 80 dB rests entirely on an uncalibrated 1 pW detector floor, and the attribution to back reflections is not directly evidenced. These issues are load-bearing for the headline result.

major comments (3)
  1. [Results and Discussion, Fig. 4] The >80 dB claim is not established. The text states that optimization "reduces the detected power to less than 1 pW which is the nominal detection limit of our photodiode" and that this "represents a maximum measurable extinction ratio of 81.8 dB." Because the self-configuration algorithm minimizes the photodiode reading directly, reaching the detector's nominal floor does not demonstrate that the optical power is below 1 pW. No dark-current subtraction, electrical-offset calibration, noise-floor measurement, detector linearity check, or repeated run is reported. The sentence in the same paragraph acknowledging that "apparent rejection beyond 81.8 dB may not be reliable" does not repair the abstract and introduction claims of "exceeding 80 dB." The authors should either provide a calibrated floor (dark-current measurement, repeated measurements, or a control experiment at different input powers) or restate the result as a detector-floor-limited upper bound.
  2. [Results and Discussion, off-path DMZI discussion] The attribution of the off-light-path device improvement to "non-negligible reflections from the coupling interfaces" is not directly evidenced. The data show only that tuning some off-path devices changes the detector reading; they do not isolate a reflection path from other mechanisms such as residual leakage through nominally transparent devices, thermal crosstalk between adjacent heaters, or stray optical coupling. Since the paper's architectural claim is that redundant off-path devices compensate back reflections, this mechanism should be tested, for example by comparing devices with different coupler designs, by varying the coherence length, or by measuring the wavelength dependence. Without such evidence, the improvement from off-path devices is an empirical observation whose physical origin is unverified.
  3. [Results and Discussion, Figs. 3 and 4] All quantitative results come from a single device and a single configuration run, with no error bars, no repeated measurements, and no device-to-device statistics. For a paper whose headline is a factor-of-10^8 extinction ratio, the absence of any repeatability information is a serious limitation. At minimum, repeated configuration runs on the same device should be reported, and preferably measurements on multiple nominally identical meshes, to show that the 80 dB level is reproducible rather than a single outlier.
minor comments (3)
  1. [Introduction / Methods, Eq. (1)] Equation (1) contains a typographical error: the lower-left matrix element should read e^{j\Delta\phi}\cos(\Delta\theta/2), not e^{j\Delta\phi}\cos(\theta/2), and the same notational issue affects adjacent terms; also "extinction ration" appears in the text.
  2. [Results and Discussion, normalization] The text says optimization of DMZI [1,1] reduces leakage "by 63.3 dB from the total input power," but Figure 4 is normalized by the maximum measured output power of 150 µW, not the 2 mW laser input. This discrepancy should be corrected or clarified, since the stated 81.8 dB floor corresponds to 150 µW / 1 pW.
  3. [Methods, test structures] The measurement description would benefit from stating the detector integration time, optical bandwidth, and whether the 1 pW figure is a manufacturer specification or a measured value; these details are needed to interpret the floor.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the >80 dB extinction claim is a direct detector-floor-limited measurement, not a fitted prediction; self-cited DMZI algorithms are independently corroborated.

full rationale

The paper's central claim is experimental: after sequential optimization, the D1 detector reads below the HP 81532A's nominal 1 pW floor, giving a maximum measurable extinction ratio of 81.8 dB relative to the 150 µW maximum system output. This is a direct measurement, not a quantity recovered from a model fitted to the same data. The configuration algorithm minimizes the D1 reading that also defines the extinction ratio, but using the null-port signal as the optimization target is standard nulling practice rather than a circular derivation: the architecture is not 'predicted' from the metric, it is evaluated by it. Self-citations (Miller 2013, 2015; Pai et al. 2019; Sun et al. 2023) supply the DMZI building block and self-configuration procedure, but these are prior published results whose effectiveness is independently evidenced by the paper's standalone DMZI measurement (62.4 dB in agreement with Wilkes et al. 2016) and by the cited experimental literature; no uniqueness theorem or ansatz is imported solely through self-citation. The paper itself flags the detector-floor limitation in Results and Discussion: 'Optimization of DMZI [2, 1] further reduces the detected power to less than 1 pW which is the nominal detection limit of our photodiode. This represents a maximum measurable extinction ratio of 81.8 dB' and later notes 'Apparent rejection beyond 81.8 dB may not be reliable because we are below the nominal detection limit.' That is a measurement-validity/correctness caveat, not a circularity in the derivation chain. Therefore no specific circular step is identified; the score reflects only the presence of non-load-bearing self-citations and the measurement-limited nature of the headline number.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

No free parameters are used in the central claim; the split-ratio error values σ are inferred from measurements and do not feed into the >80 dB result. The central claim rests on measurement assumptions about the detector floor and on an unverified reflection mechanism.

assumptions (4)
  • domain assumption Directional couplers can be described by the split-ratio error model B(σ) in Eq. (2), and phase heaters can compensate phase errors.
    The DMZI compensation model in Methods assumes fabrication errors appear only as split-ratio errors σ, with no other parasitic effects such as polarization or wavelength dependence.
  • domain assumption The photodiode's 1 pW nominal minimum detectable power bounds the true optical power at the null port.
    The >80 dB claim is based on the detector floor in Results and Discussion; no dark current calibration or noise floor measurement is reported.
  • ad hoc to paper The improvement from off-light-path DMZIs is caused by back reflections from coupling interfaces.
    The authors write 'We attribute these effects to non-negligible reflections from the coupling interfaces,' but no direct reflection measurement or simulation is provided.
  • domain assumption The sequential configuration algorithm converges to an optimal state for each DMZI that is stable across the mesh.
    The Methods algorithm is stated as converging, but convergence criteria and thermal crosstalk characterization are not reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of High Contrast Nulling in Photonic Meshes Through Architectural Redundancy." pith.science (2026). https://pith.science/paper/HQSGQ67T

@misc{pith2026250207997,
  author       = {Pith},
  title        = {Pith review of: High Contrast Nulling in Photonic Meshes Through Architectural Redundancy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HQSGQ67T}},
  note         = {Machine review of arXiv:2502.07997}
}
read the original abstract

We demonstrate a silicon photonic architecture comprised of Double Mach-Zehnder Interferometers (DMZIs) designed for high-contrast photonic applications. This configuration significantly enhances the achievable extinction ratio of photonic integrated circuits (PICs), reaching levels exceeding 80 dB. By leveraging the tunable properties of DMZIs and implementing a systematic configuration algorithm, the proposed mesh effectively compensates for fabrication imperfections and mitigates non-idealities such as back reflections. Experimental validation on a silicon-on-insulator platform demonstrates the potential of this approach for applications requiring high contrast nulling such as astronomical sensing.

Figures

Figures reproduced from arXiv: 2502.07997 by the authors.

Figure 1
Figure 1. (a). Schematic diagram of an ideal Mach-Zehnder inter￾ferometer with perfect split ratios. Phase shifters ∆θ and ∆ϕ set the matrix elements of the 2x2 transformation applied to input fields E1 and E2. The output fields are measured at de￾tectors D1 and D2. (b). Schematic diagram of a Double Mach￾Zehnder Interferometer with tunable beam splitters. Phase shifters ∆θL and ∆θR control the split ratio of the left and rig… view at source ↗
Figure 2
Figure 2. Schematic diagram of a 3x3 triangular mesh where each node is a Double Mach-Zehnder Interferometer. Subsequent elements of a single diagonal line along the light path are used to successively filter leakage from the previous elements. Ad￾ditionally elements not along the light path are configured to reduce the impact of back reflections. We use a fiber coupled HP 81680A tunable C-band laser pro￾viding 2 mW of optica… view at source ↗
Figure 3
Figure 3. Output power at detector D1 as heater power to phase shifter ∆θ in both the standard MZI and the DMZI is swept. The measured extinction ratio for a standard MZI is 42.0 dB corresponding to a split ratio 49.6 : 50.4 at each of the beam splitters. The measured extinction ratio for a configured DMZI is 62.4 dB corresponding to a split ratio of 49.96 : 50.04 at each of the tunable beam splitters. The results of this swe… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: (a).Relative detected power at the output photodiode D1 of the 3x3 triangular mesh as the internal phase shifter ∆θ of each of the 6 devices is tuned sequentially. (b). After each device has been optimized, the optimal phase shifter settings are held steady while the s…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Low-loss, fabrication-tolerant, and highly-tunable Sagnac loop reflectors and Fabry-P\'erot cavities on thin-film lithium niobate

    physics.optics 2025-05 conditional novelty 6.0 of 10

    Tunable Sagnac-loop-based Fabry-Perot cavities on thin-film lithium niobate yield an inferred per-MZI loss below 1.5% and a thermo-optic phase-shift power of 2.5 mW.

Reference graph

Works this paper leans on

46 extracted references · 46 canonical work pages · cited by 1 Pith paper

  1. [1]

    Z. Wen, Z. Guan, J. Dong, et al. , Sensors (Basel, Switzerland) 22 (2022)

    work page 2022

  2. [2]

    Huang, G

    T. Huang, G. Xu, X. Tu, et al., J. Opt. Soc. Am. B 38, 2680 (2021)

    work page 2021

  3. [3]

    Cho and T.-G

    S.-B. Cho and T.-G. Noh, Opt. Express 17, 19027 (2009)

    work page 2009

  4. [4]

    A. Hati, C. Nelson, J. Taylor,et al., Photonics Technol. Lett. IEEE 20, 1842 (2008)

    work page 2008

  5. [5]

    Bogaerts, D

    W. Bogaerts, D. Pérez, J. Capmany, et al., Nature 586, 207 (2020)

    work page 2020

  6. [6]

    Perez, I

    D. Perez, I. Gasulla, J. Fraile-Pelaez,et al., Laser & Photonics Rev. 11 (2017)

    work page 2017

  7. [7]

    Bogaerts and L

    W. Bogaerts and L. Chrostowski, Laser & Photonics Rev. 12, 1700237 (2018)

    work page 2018

  8. [8]

    P . C. Humphreys, B. J. Metcalf, J. B. Spring,et al., Opt. Express 22, 21719 (2014)

    work page 2014

Show all 46 references
  1. [9]

    Enright, S

    R. Enright, S. Lei, K. Nolan, et al., Bell Labs Tech. J. 19, 31 (2014)

    2014

  2. [10]

    Z. Sun, S. Pai, C. Valdez, et al., Optica 10, 1165 (2023)

    2023

  3. [11]

    S. Pai, T. Park, M. Ball, et al., Optica 10, 552 (2023)

    2023

  4. [12]

    Cheng, H

    J. Cheng, H. Zhou, and J. Dong, Nanomaterials 11 (2021)

    2021

  5. [13]

    Annoni, E

    A. Annoni, E. Guglielmi, M. Carminati, et al. , Light. Sci. & Appl. 6, e17110 (2017)

    2017

  6. [14]

    Milanizadeh, S

    M. Milanizadeh, S. SeyedinNavadeh, F . Zanetto,et al., Light. Sci. & Appl. 11, 197 (2022)

    2022

  7. [15]

    Integrated photonic-based coronagraphic systems for future space telescopes,

    N. Desai and Et al., “Integrated photonic-based coronagraphic systems for future space telescopes,” in Proc.SPIE, , vol. 12680 (2023), p. 126801S

    2023

  8. [16]

    AstroPIC: Architecture options and trades for integrated photonic coronagraphy,

    D. Sirbu, R. Belikov, K. Fogarty,et al., “AstroPIC: Architecture options and trades for integrated photonic coronagraphy,” inAmerican Astro- nomical Society Meeting Abstracts, , vol. 243 of American Astronomical Society Meeting Abstracts (2024), p. 329.07

    2024

  9. [17]

    S. Pai, B. Bartlett, O. Solgaard, and D. A. B. Miller, Phys. Rev. Appl.11, 064044 (2019)

    2019

  10. [18]

    D. A. B. Miller, Optica 2, 747 (2015)

    2015

  11. [19]

    C. M. Wilkes, X. Qiang, J. Wang, et al., Opt. Lett. 41, 5318 (2016)

    2016

  12. [20]

    M. Wang, A. Ribero, Y . Xing, and W. Bogaerts, Opt. Express28, 5555 (2020)

    2020

  13. [21]

    D. A. B. Miller, Photon. Res. 1, 1 (2013)

    2013

  14. [22]

    Hamerly, S

    R. Hamerly, S. Bandyopadhyay, and D. Englund, Nat. Commun. 13, 6831 (2022)

    2022

  15. [23]

    Suzuki, G

    K. Suzuki, G. Cong, K. Tanizawa, et al., Opt. Express 23, 9086 (2015). Letter 5 FULL REFERENCES

    2015

  16. [24]

    A review of sensitivity enhancement in interferometer-based fiber sensors,

    Z. Wen, Z. Guan, J. Dong, et al., “A review of sensitivity enhancement in interferometer-based fiber sensors,” Sensors (Basel, Switzerland) 22 (2022)

    2022

  17. [25]

    Design of highly sensitive interferometric sensors based on subwavelength grating waveguides operating at the dispersion turning point,

    T. Huang, G. Xu, X. Tu,et al., “Design of highly sensitive interferometric sensors based on subwavelength grating waveguides operating at the dispersion turning point,” J. Opt. Soc. Am. B 38, 2680–2686 (2021)

    2021

  18. [26]

    Stabilization of a long-armed fiber-optic single-photon interferometer,

    S.-B. Cho and T.-G. Noh, “Stabilization of a long-armed fiber-optic single-photon interferometer,” Opt. Express17, 19027–19032 (2009)

    2009

  19. [27]

    Cancellation of vibration-induced phase noise in optical fibers,

    A. Hati, C. Nelson, J. Taylor,et al., “Cancellation of vibration-induced phase noise in optical fibers,” Photonics Technol. Lett. IEEE20, 1842– 1844 (2008)

    2008

  20. [28]

    Programmable photonic circuits,

    W. Bogaerts, D. Pérez, J. Capmany, et al., “Programmable photonic circuits,” Nature 586, 207–216 (2020)

    2020

  21. [29]

    Silicon photonics rectan- gular universal interferometer,

    D. Perez, I. Gasulla, J. Fraile-Pelaez,et al., “Silicon photonics rectan- gular universal interferometer,” Laser & Photonics Rev.11 (2017)

    2017

  22. [30]

    Silicon photonics circuit design: Methods, tools and challenges,

    W. Bogaerts and L. Chrostowski, “Silicon photonics circuit design: Methods, tools and challenges,” Laser & Photonics Rev. 12, 1700237 (2018)

    2018

  23. [31]

    Strain-optic active control for quantum integrated photonics,

    P . C. Humphreys, B. J. Metcalf, J. B. Spring, et al., “Strain-optic active control for quantum integrated photonics,” Opt. Express 22, 21719 (2014)

    2014

  24. [32]

    A vision for thermally integrated photonics systems,

    R. Enright, S. Lei, K. Nolan, et al., “A vision for thermally integrated photonics systems,” Bell Labs Tech. J.19, 31–45 (2014)

    2014

  25. [33]

    Scalable low-latency optical phase sensor array,

    Z. Sun, S. Pai, C. Valdez, et al., “Scalable low-latency optical phase sensor array,” Optica10, 1165–1172 (2023)

    2023

  26. [34]

    Experimental evaluation of digitally veri- fiable photonic computing for blockchain and cryptocurrency,

    S. Pai, T. Park, M. Ball,et al., “Experimental evaluation of digitally veri- fiable photonic computing for blockchain and cryptocurrency,” Optica 10, 552–560 (2023)

    2023

  27. [35]

    Photonic matrix computing: From fundamentals to applications,

    J. Cheng, H. Zhou, and J. Dong, “Photonic matrix computing: From fundamentals to applications,” Nanomaterials 11 (2021)

    2021

  28. [36]

    Unscrambling light— automatically undoing strong mixing between modes,

    A. Annoni, E. Guglielmi, M. Carminati, et al. , “Unscrambling light— automatically undoing strong mixing between modes,” Light. Sci. & Appl. 6, e17110–e17110 (2017)

    2017

  29. [37]

    Separating arbitrary free-space beams with an integrated photonic processor,

    M. Milanizadeh, S. SeyedinNavadeh, F . Zanetto, et al. , “Separating arbitrary free-space beams with an integrated photonic processor,” Light. Sci. & Appl. 11, 197 (2022)

    2022

  30. [38]

    Integrated photonic-based coronagraphic sys- tems for future space telescopes,

    N. Desai and Et al., “Integrated photonic-based coronagraphic sys- tems for future space telescopes,” in Proc.SPIE, vol. 12680 (2023), p. 126801S

    2023

  31. [39]

    AstroPIC: Architecture options and trades for integrated photonic coronagraphy,

    D. Sirbu, R. Belikov, K. Fogarty,et al., “AstroPIC: Architecture options and trades for integrated photonic coronagraphy,” inAmerican Astro- nomical Society Meeting Abstracts, vol. 243 of American Astronomical Society Meeting Abstracts (2024), p. 329.07

    2024

  32. [40]

    Matrix optimization on universal unitary photonic devices,

    S. Pai, B. Bartlett, O. Solgaard, and D. A. B. Miller, “Matrix optimization on universal unitary photonic devices,” Phys. Rev. Appl. 11, 064044 (2019)

    2019

  33. [41]

    Perfect optics with imperfect components,

    D. A. B. Miller, “Perfect optics with imperfect components,” Optica 2, 747–750 (2015)

    2015

  34. [42]

    60db high-extinction auto- configured mach–zehnder interferometer,

    C. M. Wilkes, X. Qiang, J. Wang, et al. , “60db high-extinction auto- configured mach–zehnder interferometer,” Opt. Lett. 41, 5318–5321 (2016)

    2016

  35. [43]

    Tolerant, broadband tunable coupler circuit,

    M. Wang, A. Ribero, Y . Xing, and W. Bogaerts, “Tolerant, broadband tunable coupler circuit,” Opt. Express 28, 5555–5566 (2020)

    2020

  36. [44]

    Self-configuring universal linear optical component,

    D. A. B. Miller, “Self-configuring universal linear optical component,” Photon. Res. 1, 1–15 (2013)

    2013

  37. [45]

    Asymptotically fault- tolerant programmable photonics,

    R. Hamerly, S. Bandyopadhyay, and D. Englund, “Asymptotically fault- tolerant programmable photonics,” Nat. Commun. 13, 6831 (2022)

    2022

  38. [46]

    Ultra-high-extinction-ratio 2× 2 silicon optical switch with variable splitter,

    K. Suzuki, G. Cong, K. Tanizawa,et al., “Ultra-high-extinction-ratio 2× 2 silicon optical switch with variable splitter,” Opt. Express23, 9086–9092 (2015)

    2015

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.