REVIEW 2 major objections 5 minor 1 cited by
Towards experimental demonstration of quantum position verification using true single photons
T0 review · 2 major / 5 minor · reviewed 2026-08-08 · deepseek-v4-flash
Pith's one-line read This paper reports an experimental implementation of a loss-tolerant quantum position verification protocol using true single photons from a quantum dot source, where the prover's SWAP measurement relies on Hong-Ou-Mandel interference…
desk verdict Solid experimental first step toward loss-tolerant QPV, but the security threshold comparison is miscalibrated because the experiment uses one basis while the 2/3 LOCC bound assumes three; must be fixed before the outlook claims are supportable. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central mechanism is the loss-tolerant SWAP measurement: two photons enter a 50:50 beamsplitter (BS1); if they are in the same polarization state, Hong-Ou-Mandel bunching sends both to the same output arm, and two more 50:50 beamsplitters (BS2, BS3) with four single-photon detectors let the prover distinguish a bunched (parallel) event from a non-bunched (orthogonal) event while keeping loss events unrecognizable as either. The protocol's security-relevant quantity is the conditional probability $P(0|\parallel,\mathrm{concl.})$ that the prover returns $z=0$ when the verifiers sent parallel qubits; the source parameters that govern it are the single-photon purity $P = 1 - g^{(2)}(0)$ and the wave-function overlap $M$ obtained from the Hong-Ou-Mandel visibility.
What would settle it
A direct test is to run the same prover with the encoding in all three mutually unbiased bases and with a source whose purity and indistinguishability match the best-case parameters considered in the paper (0.979 and 0.960): if the measured conditional probability for parallel qubits is at or below 2/3, the paper's projection that an improved source would yield secure QPV is refuted.
Extended reading notes
Core claim
The paper establishes that a loss-tolerant QPV protocol can be run with true single photons: using a demultiplexed quantum-dot source, two verifiers send polarization qubits to a prover who performs the SWAP measurement — two-photon Hong-Ou-Mandel interference at beamsplitter BS1 followed by two additional beamsplitters and four detectors that discriminate bunching from loss. For orthogonal qubits, the measured conditional probabilities match the ideal expectation ($P(0|\perp,\mathrm{concl.}) = 0.34$ versus $1/3$), but for parallel qubits the measured $P(0|\parallel,\mathrm{concl.}) = 0.48$ falls far short of the ideal value 1 because the source's single-photon purity ($P = 0.776$) and wave-function overlap ($M = 0.542$) are too low. Modeling the experiment with the source parameters of a state-of-the-art quantum dot source (purity 0.979, indistinguishability 0.960) predicts $P(0|\parallel,\mathrm{concl.}) \approx 0.87$, which would exceed the 2/3 threshold that limits LOCC adversaries; the conclusion is that the current setup cannot yet claim fully secure QPV, but the bottleneck is identified as a source problem that is in principle avoidable.
Load-bearing premise
The security-relevant comparison assumes the 2/3 LOCC bound applies, but that bound holds only when the verifiers use all three mutually unbiased polarization bases, whereas the experiment encodes only in the horizontal–vertical basis.
Editorial extensions
If this is right
- The experiment shows that a loss-tolerant QPV prover can be built from standard fiber components and a quantum-dot single-photon source, so the main remaining obstacle is source quality, not protocol design.
- With a source matching the best-case parameters considered in the paper (purity 0.979, indistinguishability 0.960), the honest prover's correct-answer probability for parallel qubits is projected to reach about 0.87, exceeding the 2/3 LOCC threshold and making the protocol secure against LOCC adversaries.
- The parameter map in the paper implies that both high purity and high indistinguishability are required; improving only one leaves the protocol below threshold.
- The slow-quantum-information loophole — light travelling slower in fiber than in free space — is not addressed here, so practical deployment in existing fiber networks awaits protocols with a commitment step.
Reading between the lines
- The paper compares its single-basis (HV) results against a 2/3 threshold that is proven for protocols using all three mutually unbiased bases; a fair security comparison would require repeating the measurement in all three bases, since an LOCC attacker who knows the basis can otherwise succeed with certainty.
- Because the same setup's polarization modulators can already prepare arbitrary states, extending the demonstration to three mutually unbiased bases appears to be a straightforward follow-up that would make the threshold comparison meaningful.
- The model's case C suggests that improving purity alone (e.g., by suppressing re-excitation and background emission) raises $P(0|\parallel,\mathrm{concl.})$ from 0.48 to about 0.59, so a modest source improvement may already bring the experiment close to threshold.
- A natural next experiment is to measure the full coincidence distribution for all three bases and extract $P(0|\parallel,\mathrm{concl.})$ per basis, which would directly test whether the LOCC bound is the right security benchmark for the implemented protocol.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper reports an experimental implementation of the loss-tolerant SWAP quantum position verification protocol using a demultiplexed quantum-dot single-photon source. Two verifiers prepare polarization qubits in the HV basis, send them through fiber delays to a prover, and the prover performs a Hong-Ou-Mandel interference measurement at a beamsplitter followed by two additional beamsplitters and four detectors. The authors measure two-fold coincidence statistics for parallel and orthogonal qubits, compare the conditional probabilities with ideal values and with a model incorporating source purity, indistinguishability, and measured setup losses, and conclude that the current source imperfections prevent fully secure QPV but that an improved source would exceed the LOCC threshold of 2/3. The central claim is that these are first results towards an experimental demonstration of quantum position verification.
Significance. If the central comparison were valid, this would be a useful experimental step: it demonstrates a loss-tolerant prover measurement with true single photons from a quantum-dot source, and it provides a quantitative model whose inputs (purity, indistinguishability, component transmissions) are independently measured rather than fitted. The data for orthogonal qubits reproduce the predicted 1/3-2/3 split, and the suppression of parallel-qubit coincidences is clearly visible and reproduced by the model at 0.47 predicted versus 0.48 measured. The main limitation is that the security-relevant LOCC threshold is not applicable to the implemented single-basis configuration, which undermines the quantitative interpretation of the threshold comparisons. The paper is honest about not yet achieving fully secure QPV and about the basis restriction, but the way the threshold is used makes the central security claim misleading as written.
major comments (2)
- [Section I, Eq. (2), Fig. 4, Table II, Fig. 5] The LOCC threshold P_LOCC = 2/3 in Eq. (2) is derived under the explicit assumption that the verifiers use all three mutually unbiased bases, as stated in the LOCC-attack paragraph. The experiment uses only the HV basis, as stated in Section I ('we show here one basis only'). For a single publicly known basis, an LOCC adversary can measure each intercepted photon in HV and compare the two outcomes; in the ideal lossless case this succeeds with probability 1, not 2/3. Therefore the dashed 'LOCC limit' in Fig. 4 is not a valid security threshold for the demonstrated configuration, and the comparisons in Table II and Fig. 5 of improved-source projections against 2/3 do not indicate progress toward LOCC security. The paper should either implement all three mutually unbiased bases, or derive and use a single-basis-specific LOCC bound, or explicitly restrict the claims to a demonstration of the honest prover's measurement without asserting that exceeding 2/3 approaches LOCC security.
- [Section IV and Table II] The model used to produce the predicted values P(0|∥,concl) = 0.47, 0.87, and 0.59 in Table II and the contour plot in Fig. 5 is described only verbally in the main text as 'a simple model of our experiment including photon source parameters, and all characteristics of the optical setup including loss, unbalanced fiber beam splitters, and detection efficiencies'. No explicit equation, algorithm, or parameterized rule is given in the main text or in the included supplemental material. Since the projection that an improved source would exceed the 2/3 threshold is a central quantitative claim of the paper, the model should be stated explicitly, or a precise reference to a derivable supplemental equation should be provided so that the reader can reproduce the projected values.
minor comments (5)
- [Section III and Fig. 3 caption] The measurement duration is inconsistent: the main text says data is recorded in 5-minute intervals and mentions 'one-hour long measurements', while the Fig. 3 caption says '5 hour long measurement'. Please reconcile these numbers.
- [Table I] The experimental values for P(⊘|⊥) and P(⊘|∥) are listed as 'NA'. Because the protocol's verification step includes inconclusive responses and the loss-tolerance argument depends on distinguishing loss from conclusive events, the authors should state explicitly whether the inconclusive fraction was measured and, if not, why it is omitted from the analysis.
- [Section IV, Eqs. (3) and (4)] In Eqs. (3) and (4), the quantities g^(2)_⊥, g^(2)_∥, and g^(2) are not fully defined: please state explicitly that these are zero-time second-order correlation functions, which HOM configuration each refers to, and which value of g^(2) enters Eq. (4).
- [Fig. 4] The dotted modeled bars should be identified in the legend rather than only in the caption, and the error bars or uncertainty ranges for the modeled values should be defined in the figure or its caption.
- [Section II and Table S1] The detector efficiencies in Table S1 are normalized to detector A, which is listed as 100%; stating the absolute efficiency of at least one detector would allow the absolute coincidence rates to be compared with the model.
Circularity Check
No circular derivation: the model inputs are independently measured and the LOCC threshold is an external theoretical bound, not an input fitted to the data.
full rationale
The paper's central experimental claim is that the prover's loss-tolerant SWAP measurement is implemented with true single photons, and the measured conditional probabilities (Table I) are compared to an independent protocol model. I find no circular step. The model inputs are independently measured: single-photon purity P from a Hanbury-Brown-Twiss g^(2) measurement, indistinguishability M from a separate Hong-Ou-Mandel visibility measurement (Eqs. 3-4), and all component transmissions and splitting ratios from the supplementary characterization (Tables S1-S2). The model output for the authors' own source, P(0|parallel,concl.) = 0.47 +/- 0.03, matches the experimentally measured 0.48 without any fitting of the QPV coincidence data to the model; this is a forward validation, not a fitted input renamed as prediction. The LOCC security threshold 2/3 used in Fig. 4 is not an input derived from the experimental data; it is stated in Eq. (2) and attributed to the authors' prior theoretical work [24]. Although Ref. [24] shares authors with the present paper, the bound is a parameter-free mathematical statement with stated assumptions (all three mutually unbiased bases) and is externally checkable; the present paper also provides a self-contained sketch of the 1/3(1+1/2+1/2) calculation. The fact that the experiment uses only the HV basis while the bound assumes three MUBs is a correctness and validity mismatch that should be addressed, but it is not circularity: the threshold is not defined in terms of the experiment's outputs. No equation in the paper reduces to its own input by construction.
Assumptions & free parameters
assumptions (5)
- domain assumption The LOCC attack success probability is bounded by 2/3 when the verifiers use all three mutually unbiased polarization bases (Eq. 2).
- domain assumption The single-photon purity P and indistinguishability M are sufficient source parameters to model the prover's coincidence probabilities.
- domain assumption Equation (4), M = V_HOM (1 + 2g(2)), correctly accounts for the effect of multi-photon emission on the HOM visibility.
- standard math The HOM bunching at BS1 deterministically sends two parallel indistinguishable photons to the same output, while orthogonal photons leave independently.
- domain assumption The SWAP protocol from Ref. [24] is correct and secure under the stated loss model, so the prover's answer rule (AB/CD = 0, others = 1) is the right one.
Cite this review
Pith. "Pith review of Towards experimental demonstration of quantum position verification using true single photons." pith.science (2026). https://pith.science/paper/YG2X3N6U
@misc{pith2026250204125,
author = {Pith},
title = {Pith review of: Towards experimental demonstration of quantum position verification using true single photons},
year = {2026},
howpublished = {\url{https://pith.science/paper/YG2X3N6U}},
note = {Machine review of arXiv:2502.04125}
}
read the original abstract
The geographical position can be a good credential for authentication of a party, this is the basis of position-based cryptography - but classically this cannot be done securely without physical exchange of a private key. However, recently, it has been shown that by combining quantum mechanics with the speed of light limit of special relativity, this might be possible: quantum position verification. Here we demonstrate experimentally a protocol that uses two-photon Hong-Ou-Mandel interference at a beamsplitter, which, in combination with two additional beam splitters and 4 detectors is rendering the protocol resilient to loss. With this we are able to show first results towards an experimental demonstration of quantum position verification.
Figures
Figures from the paper (3 more)
Forward citations
Cited by 1 Pith paper
-
Efficient Multi-basis Quantum Position Verification Secure against Generalized Adversaries
A composed QKD plus quantum position verification protocol authenticates parties by location, with a generalized adversary analysis and a new multi-basis QPV scheme.
Reference graph
Works this paper leans on
-
[24]
Junge, M., Kubicki, A. M., Palazuelos, C. & Pérez- García, D. Geometry of Banach Spaces: A New Route Towards Position Based Cryptography.Communications in Mathematical Physics 394, 625 (2022)
work page 2022
-
[1]
Encoded in single photons, these qubits are sent to the prover such that they arrive simultaneously
Preparation: Verifiers V0 and V1 share via their private channel a uniformly drawn random se- quence of basis choices and randomly parallel or orthogonal states in the basis, e.g.|Ψ0⟩ and|Ψ1⟩. Encoded in single photons, these qubits are sent to the prover such that they arrive simultaneously
-
[2]
Measurement ⋆: The prover performs the quan- tum measurement based on two-photon Hong-Ou- Mandel (HOM) quantum interference [31] but with twoadditionalbeamsplittersand4detectors, which allows to discriminate HOM photon bunching from loss as explained below. The prover returns a clas- sical response z = 0 if |Ψ0⟩ ∥ |Ψ1⟩, z = 1 if |Ψ0⟩ ⊥ |Ψ1⟩, and z = ⊘ if ...
-
[3]
If either check fails the verifiers abort the protocol
Round check: After each response of the prover the verifiers review if the received responsez is the same for both verifiers and if the response arrived within the set time constraint. If either check fails the verifiers abort the protocol
-
[4]
Verification: After n rounds of steps 1 . . .3 the verifiers check if the distributions of answers re- turned by the proverz ={0, 1, ⌀} follows the ex- pected distribution within a certain error margin. II. EXPERIMENT A. The single-photon source Essential for our experiment shown in Fig. 2 is the source of single photons. We use a single negatively charge...
-
[5]
Chandran, N., Goyal, V., Moriarty, R. & Ostrovsky, R. Position Based Cryptography. In Halevi, S. (ed.) Advances in Cryptology - CRYPTO 2009 , 391–407 (Springer, Berlin, Heidelberg, 2009)
work page 2009
-
[6]
Kent, A., Munro, W., Spiller, T. & Beausoleil, R. Tag- ging systems, 2006. US patent (2006)
work page 2006
-
[7]
Kent, A., Munro, W. J. & Spiller, T. P. Quantum tag- ging: Authenticating location via quantum information and relativistic signaling constraints.Physical Review A 84, 012326 (2011)
work page 2011
Show all 45 references
-
[8]
Quantum tagging for tags containing secret classical data
Kent, A. Quantum tagging for tags containing secret classical data. Physical Review A 84, 022335 (2011)
2011
-
[9]
The conundrum of secure positioning.Na- ture 479, 307 (2011)
Brassard, G. The conundrum of secure positioning.Na- ture 479, 307 (2011)
2011
-
[10]
Quantum position verification in the random oracle model (2014)
Unruh, D. Quantum position verification in the random oracle model (2014). Cryptoeprint:2014/118
2014
-
[11]
& Beausoleil, R
Kent, A., Munro, W., Spiller, T. & Beausoleil, R. Tag- ging Systems (2006)
2006
-
[12]
Malaney, R. A. Location-Dependent Communications Using Quantum Entanglement.Phys. Rev. A 81, 042319 (2010)
2010
-
[13]
Malaney, R. A. Quantum Location Verification in Noisy Channels. In 2010 IEEE Global Telecommunications Conference GLOBECOM 2010, 1–6 (2010). 1004.4689
2010 arXiv
-
[14]
& Schaffner, C
Buhrman, H., Chandran, N., Fehr, S., Gelles, R., Goyal, V., Ostrovsky, R. & Schaffner, C. Position-Based Quan- tum Cryptography: Impossibility and Constructions. SIAM J. Comput. 43, 150 (2014)
2014
-
[15]
& Siopsis, G
Qi, B. & Siopsis, G. Loss-Tolerant Position-Based Quan- tum Cryptography. Phys. Rev. A 91, 042337 (2015). 1502.02020
2015 arXiv
-
[16]
Miller, C. A. & Alnawakhtha, Y. Perfect cheating is impossible for single-qubit position verification (2024). ArXiv:2406.20022 [quant-ph]
2024 arXiv
-
[17]
& Pistoia, M
Amer, O., Chakraborty, K., Cui, D., Kaleoglu, F., Lim, C., Liu, M. & Pistoia, M. Certified Random- ness implies Secure Classical Position-Verification (2024). ArXiv:2410.03982 [quant-ph]
2024 arXiv
-
[18]
Escolà-Farràs, L., Palais, L. C. & Speelman, F. A quan- tum cloning game with applications to quantum position verification (2024). ArXiv:2410.22157 [quant-ph]
2024
-
[19]
George, I., Allerstorfer, R., Lunel, P. V. & Chitambar, E. Orthogonality Broadcasting and Quantum Position Verification (2024). ArXiv:2311.00677 [quant-ph]
2024
-
[20]
Quantum tasks in holography.Journal of High Energy Physics 2019, 233 (2019)
May, A. Quantum tasks in holography.Journal of High Energy Physics 2019, 233 (2019)
2019
-
[21]
& Grosshans, F
Olivo, A., Chabaud, U., Chailloux, A. & Grosshans, F. Breaking simple quantum position verification proto- cols with little entanglement (2020). ArXiv:2007.15808 [quant-ph]
2020 arXiv
-
[22]
& Siopsis, G
Das, S. & Siopsis, G. Practically secure quantum position verification. New Journal of Physics 23, 063069 (2021)
2021
-
[23]
BeatingClassicalImpossibility of Position Verification (2022)
Liu, J., Liu, Q.&Qian, L. BeatingClassicalImpossibility of Position Verification (2022). ArXiv:2109.07517 [quant- ph]
2022 arXiv
-
[25]
& May, A
Cree, J. & May, A. Code-routing: a new attack on posi- tion verification. Quantum 7, 1079 (2023)
2023
-
[26]
& Verduyn Lunel, P
Allerstorfer, R., Buhrman, H., May, A., Speelman, F. & Verduyn Lunel, P. Relating non-local quantum compu- tation to information theoretic cryptography.Quantum 8, 1387 (2024)
2024
-
[27]
& Lunel, P
Allerstorfer, R., Buhrman, H., Speelman, F. & Lunel, P. V. On the Role of Quantum Communication and Loss in Attacks on Quantum Position Verification (2022). 2208.04341
2022 arXiv
-
[28]
& Lunel, P
Allerstorfer, R., Buhrman, H., Speelman, F. & Lunel, P. V. Towards Practical and Error-Robust Quantum Po- sition Verification (2022). 2106.12911
2022 arXiv
-
[29]
Lim, C. C. W., Xu, F., Siopsis, G., Chitambar, E., Evans, P. G. & Qi, B. Loss-Tolerant Quantum Secure Position- ing with Weak Laser Sources.Phys. Rev. A 94, 032315 (2016)
2016
-
[30]
& Speelman, F
Bluhm, A., Christandl, M. & Speelman, F. A Single- Qubit Position Verification Protocol That Is Secure 7 against Multi-Qubit Attacks.Nat. Phys. 18, 623 (2022)
2022
-
[31]
& May, A
Asadi, V., Cleve, R., Culf, E. & May, A. Linear gate bounds against natural functions for position-verification (2024). ArXiv:2402.18648 [quant-ph]
2024 arXiv
-
[32]
& Speelman, F
Escolà-Farràs, L. & Speelman, F. Single-Qubit Loss- Tolerant Quantum Position Verification Protocol Secure against Entangled Attackers. Physical Review Letters 131, 140802 (2023)
2023
-
[33]
& Speelman, F
Escolà-Farràs, L. & Speelman, F. Lossy-and-Constrained Extended Non-Local Games with Applications to Cryp- tography: BC, QKD and QPV (2024). ArXiv:2405.13717 [quant-ph]
2024 arXiv
-
[34]
& Lunel, P
Allerstorfer, R., Bluhm, A., Buhrman, H., Christandl, M., Escolà-Farràs, L., Speelman, F. & Lunel, P. V. Making Existing Quantum Position Verification Proto- cols Secure Against Arbitrary Transmission Loss (2023). 2312.12614
2023
-
[35]
K., Ou, Z
Hong, C. K., Ou, Z. Y. & Mandel, L. Measurement of Subpicosecond Time Intervals between Two Photons by Interference. Phys. Rev. Lett. 59, 2044 (1987)
1987
-
[36]
C., Almeida, M
Somaschi, N., Giesz, V., De Santis, L., Loredo, J. C., Almeida, M. P., Hornecker, G., Portalupi, S. L., Grange, T., Antón, C., Demory, J., Gómez, C., Sagnes, I., Lanzillotti-Kimura, N. D., Lemaítre, A., Auffeves, A., White, A. G., Lanco, L. & Senellart, P. Near-Optimal Single-...
2016
-
[37]
J., Frey, J
Snijders, H. J., Frey, J. A., Norman, J., Flayac, H., Savona, V., Gossard, A. C., Bowers, J. E., van Exter, M. P., Bouwmeester, D. & Löffler, W. Observation of the Unconventional Photon Blockade. Phys. Rev. Lett. 121, 043601 (2018)
2018
-
[38]
C., Korsch, A
Tomm, N., Javadi, A., Antoniadis, N.O., Najer, D., Löbl, M. C., Korsch, A. R., Schott, R., Valentin, S. R., Wieck, A. D., Ludwig, A. & Warburton, R. J. A Bright and Fast Source of Coherent Single Photons. Nat. Nanotechnol. 16, 399 (2021)
2021
-
[39]
S., Nilsson, M., Ciaccia, C., Jünger, C., Rossi, F., Zannier, V., Sorba, L., Baumgartner, A
Thomas, F. S., Nilsson, M., Ciaccia, C., Jünger, C., Rossi, F., Zannier, V., Sorba, L., Baumgartner, A. & Schönenberger, C. Spectroscopy of the Local Density of States in Nanowires Using Integrated Quantum Dots. Phys. Rev. B 104, 115415 (2021)
2021
-
[40]
A., Norman, J., Gos- sard, A
Steindl, P., Snijders, H., Westra, G., Hissink, E., Iakovlev, K., Polla, S., Frey, J. A., Norman, J., Gos- sard, A. C., Bowers, J. E., Bouwmeester, D. & Löffler, W. Artificial Coherent States of Light by Multiphoton Interference in a Single-Photon Stream.Phys. Rev. Lett. 126, ...
2021
-
[41]
& Löffler, W
Steindl, P., Van Der Ent, T., Van Der Meer, H., Frey, J., Norman, J., Bowers, J., Bouwmeester, D. & Löffler, W. Resonant Two-Laser Spin-State Spectroscopy of a Negatively Charged Quantum-Dot–Microcavity System with a Cold Permanent Magnet.Phys. Rev. Applied 20, 014026 (2023)
2023
-
[42]
& Löffler, W
Poortvliet, M., Steindl, P., Kuijf, I., Visser, H., van Amersfoort, A. & Löffler, W. Picosecond Laser Pulses for Quantum Dot-Microcavity Based Single Photon Genera- tion by Cascaded Electro-Optic Modulation of a Narrow- Linewidth Laser (2024). 2408.08213
2024 arXiv
-
[43]
& Löffler, W
Steindl, P., Frey, J., Norman, J., Bowers, J., Bouwmeester, D. & Löffler, W. Cross-Polarization- Extinction Enhancement and Spin-Orbit Coupling of Light for Quantum-Dot Cavity Quantum Electrodynam- ics Spectroscopy. Phys. Rev. Applied 19, 064082 (2023)
2023
-
[44]
& Pan, J.-W
Ding, X., Guo, Y.-P., Xu, M.-C., Liu, R.-Z., Zou, G.-Y., Zhao, J.-Y., Ge, Z.-X., Zhang, Q.-H., Liu, H.-L., Wang, L.-J., Chen, M.-C., Wang, H., He, Y.-M., Huo, Y.-H., Lu, C.-Y. & Pan, J.-W. High-efficiency single-photon sourceabovetheloss-tolerantthresholdforefficientlinear opt...
2023 arXiv
-
[45]
B., Bennett, A
Patel, R. B., Bennett, A. J., Cooper, K., Atkinson, P., Nicoll, C. A., Ritchie, D. A. & Shields, A. J. Postselective Two-Photon Interference from a Continuous Nonclassical Stream of Photons Emitted by a Quantum Dot.Phys. Rev. Lett. 100, 207405 (2008). 8 SUPPLEMENT AL INFORMA T...
2008
Reviewed August 8, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.