Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Evaluation of quantum key distribution systems against injection-locking attacks

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper establishes an experimental method to measure injection-locking-induced phase de-randomization in QKD transmitters, yielding a minimum required isolation of about 140 dB for the tested DFB lasers.

desk verdict Useful black-box method for injection-locking evaluation; treat the 140 dB figure as a data point for one laser pair, not as a security bound. read the letter →

arxiv 2412.10290 v1 pith:4TOTKQ76 submitted 2024-12-13 quant-ph

Jerome Wiesemann , Fadri Grünenfelder , Ana Blázquez , Nino Walenta , Davide Rusca This is my paper · ORCID
classification quant-ph
keywords quantumkeydistributioninjectionlockingphaserandomizationside-channelattackheterodynedetectiondecoy-stateBB84laserisolationblack-boxtesting
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

Practical quantum key distribution systems rely on the assumption that Alice's laser pulses have uniformly random phases, but an attacker can inject light into the transmitter to lock those phases and break the assumption. This paper develops an experimental procedure to measure how much such an injection-locking attack degrades phase randomization, using a heterodyne detection setup in which a master laser plays the role of Eve and a gain-switched slave laser plays the role of Alice. The measured quantity is a relative $q$-parameter, $q_{\rm rel}$, a lower bound on the probability density of the phase of each pulse relative to the injected light; the authors report $q^{\min}_{\rm rel}$ as a function of injected optical power. For the DFB lasers used here, injected power below about $-90$ dBm leaves phase randomization effectively unchanged, which translates into a required attenuation of roughly $140$ dB assuming Eve can inject up to $100$ W before damaging the fiber. The procedure is source-agnostic and intended for black-box evaluation of QKD transmitters, with the caveat that the result is only as strong as the master laser's ability to simulate Eve's best possible attack.

What carries the argument

The load-bearing object is the relative $q$-parameter $q_{\rm rel}$, a lower bound on the probability density of each pulse's phase relative to the master laser's reference phase, defined so that $q_{\rm rel}=1$ for perfect uniform randomization and $q_{\rm rel}\to 0$ for a fully localized phase. It is extracted from a heterodyne detection setup: a continuous-wave DFB master laser injects light into a gain-switched DFB slave laser, a $90^\circ$ optical hybrid produces in-phase and quadrature signals, and the two-argument arctangent of those signals yields the per-pulse relative phase; a histogram of $N=8000$ phases is fit with a wrapped Voigt profile to obtain $q_{\rm rel}(\tau_n)$, and $q^{\min}_{\rm rel}=\min_{\tau_n} q_{\rm rel}(\tau_n)$ is the reported figure of merit. This quantity bridges experiment and security proof: current proofs for imperfect phase randomization require either the full phase density or a parameter like $q$, so measuring $q^{\min}_{\rm rel}$ as a function of injected power turns a device vulnerability into a concrete isolation requirement. The polarization of the injected light is optimized either by scanning the Poincaré sphere for the smallest $q^{\min}_{\rm rel}$, or by minimizing back-reflected light from the unpowered slave cavity, a shortcut that works for DFB lasers and shortens the optimization from hours to minutes.

What would settle it

Repeat the power scan of Fig. 7 using a master laser with a substantially narrower linewidth or a higher-power, modulated source and check whether $q^{\min}_{\rm rel}$ falls measurably below the measured curve at any injected power; a drop, especially near $-90$ dBm, would show that the reported isolation bound is not the worst case for a real attacker.

Watch

Extended reading notes

Core claim

The central claim is that the degree of phase de-randomization caused by injection locking can be measured directly and converted into a minimum optical isolation requirement. The authors define $q_{\rm rel}$ by requiring the conditional phase density to satisfy $f(\Delta\theta^{(n)}|\Delta\theta^{(n-1)}\dots\Delta\theta^{(1)}) \ge q_{\rm rel}/(2\pi)$, so a uniform phase distribution gives $q_{\rm rel}=1$ and a phase-localized distribution gives values approaching zero. A heterodyne receiver with a $90^\circ$ optical hybrid records $I_0$ and $I_{\pi/2}$ for each pulse, and $\arctan 2(I_0,I_{\pi/2})$ gives the phase relative to the master laser; histograms of $N=8000$ pulses are fit with a wrapped Voigt profile to extract $q_{\rm rel}(\tau_n)$, and the minimum over the pulse window defines $q^{\min}_{\rm rel}$. Scanning injected power at the optimal polarization, the authors find that $q^{\min}_{\rm rel}$ stays near $0.96$ without injection and degrades smoothly as the injected power rises, with no significant influence below about $-90$ dBm; combining this threshold with a $100$ W fiber damage limit yields at least about $140$ dB of attenuation. The method also provides two polarization-optimization routes, a full Poincaré-sphere scan and a faster back-reflection measurement, which agree for the DFB lasers tested. In a black-box evaluation, the measured $q^{\min}_{\rm rel}$ is compared with the value assumed by the security proof, and since $q^{\min}_{\rm rel}\ge q$, exceeding the implementer's claimed value certifies the transmitter against this attack.

Load-bearing premise

The whole method stands on the assumption that the laboratory master laser can imitate Eve's best possible injection-locking source; if a real attacker's light couples to Alice's laser more effectively than the master laser does, the true de-randomization is stronger and the required isolation is higher than measured.

Editorial extensions

If this is right

  • A complete QKD transmitter can be put through the same test in place of the slave laser, and its measured $q^{\min}_{\rm rel}$ can be checked against the phase-randomization value assumed in the security proof.
  • For the DFB lasers measured in this paper, an isolation chain providing less than about $140$ dB of attenuation leaves the transmitter vulnerable to a $100$ W injection-locking attacker.
  • Knowing the smallest $q_{\rm rel}$ Eve can induce covers even pulse-to-pulse variation of the attack, because the bound in Eq. (4) holds for every conditional phase density.
  • The two polarization-optimization methods give equivalent results for DFB lasers, allowing fast characterization, while the Poincaré-sphere scan remains applicable to more complex transmitters with polarization-dependent losses.
  • A faster oscilloscope or a different histogram model would extend the usable time window and tighten the $q^{\min}_{\rm rel}$ estimate, which the authors identify as an experimental limitation.

Reading between the lines

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

  • If this procedure is adopted for certification, the quantitative $140$ dB number should not be read as a universal safety margin; it is specific to the tested devices and to the master laser's ability to approximate the worst realistic attacker.
  • A natural stress test would repeat the scan with a master laser that has a narrower linewidth, higher power, or modulation synchronized to the gain-switching; a drop in $q^{\min}_{\rm rel}$ at lower injected powers would show that the reported threshold is not worst-case.
  • The time-resolved nature of $q_{\rm rel}(\tau_n)$ suggests that security proofs treating the phase as constant over the pulse may need to be extended to intra-pulse phase variation, which the paper notes is absent from current analyses.
  • The same heterodyne methodology could be reused, with a suitable reference source, to characterize other phase-related side channels, such as Trojan-horse attacks that leave a detectable phase imprint in the transmitted pulses.
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 / 5 minor

Summary. This paper reports an experimental method to characterize the phase de-randomization of a gain-switched slave laser under injection locking from a master laser, using a heterodyne detection setup and a qrel metric adapted from Refs. [21,27]. The authors measure phase histograms with and without injection, optimize the injected polarization, and extract qmin_rel as a function of injected optical power. For their DFB lasers, qmin_rel remains close to the no-injection value below approximately -90 dBm; assuming a 100 W fiber damage threshold, they conclude that about 140 dB of isolation is required to protect against the attack. The paper also discusses how the method could be used for black-box testing of complete QKD transmitters.

Significance. The work addresses a real gap: experimental evaluation of injection-locking attacks on QKD phase randomization. The polarization-optimization procedure and the pulse-resolved heterodyne characterization are useful building blocks for certification workflows, and the authors are commendably explicit that the characterization is limited by the master laser's ability to emulate Eve. The bootstrap error bars and the detailed apparatus description are also strengths. However, the central quantitative claim, stated as a lower bound on the required isolation of about 140 dB, is not supported as stated: the experiment can only establish a bound for the particular master laser and slave laser used, and the reported qmin_rel values are minima of a fitted wrapped Voigt model rather than statistically certified lower bounds on the true phase distribution. The qualitative observation that the phase distribution localizes under strong injection and remains flat below about -90 dBm is convincing, but the security-oriented conclusions need re-scoping and additional conservative analysis.

major comments (3)
  1. [Abstract; Sec. IIIE; Sec. IV] The central claim that 'at least about 140 dB' of isolation is required is logically inverted relative to the experiment. The data show that, for this particular master laser, 140 dB of attenuation keeps the injected power below -90 dBm, at which the measured qmin_rel is close to the no-injection value; this demonstrates that 140 dB suffices for this Eve model, i.e., it gives an upper bound on the required isolation for that model, not a lower bound. Since a better-matched or otherwise stronger Eve source could induce stronger de-randomization at the same injected power, the experiment cannot establish a lower bound on the required isolation for an arbitrary adversary. The authors acknowledge this in Sec. IIIA ('the characterization may only be as good as the ML simulates Eve's best possible attack') and in Sec. IV, but the abstract and Sec. IIIE nevertheless present the result as a lower bound and as 'requires at least about 140 dB.' The claims should be re-scoped to a characterization for the tested ML/DUT pair, or the logical direction should be fixed.
  2. [Sec. IIIC, Eqs. (12)-(14); App. D; Fig. 7] The reported qmin_rel is the minimum of a wrapped Voigt profile fitted to each histogram, but the definition in Eq. (14) requires a lower bound on the true PDF. Least-squares fitting does not guarantee f_w ≤ f_true; a model that is too narrow or that overestimates the minimum would inflate qmin_rel and therefore understate the required isolation. The bootstrap error bars in Fig. 7 only quantify sampling uncertainty of the fitted parameters, not model misspecification. In addition, the time window used for qmin_rel is selected in App. D based on the goodness of fit, which can bias the result toward time regions where the model happens to perform well. Since the paper proposes the method for certification-style evaluation, the qmin_rel values should be conservative lower bounds, obtained either from a validated model with a safety margin or from distribution-free concentration inequalities, as the authors themselves mention in Sec. IIIC.
  3. [Sec. IIC; Sec. IIIB] The security interpretation of qmin_rel is weakened by the inequality qrel ≥ q stated in Sec. IIIB. A high measured qmin_rel implies only an upper bound on the q-parameter that enters the security proofs of Refs. [21,27]; it does not by itself certify that Alice's phase distribution is close to uniform. In the present experiment, the no-injection value q≈0.963 supports the assumption of good intrinsic randomization, but for black-box testing of an unknown transmitter, a high qrel could in principle be produced by a non-uniform Alice phase combined with master-laser phase fluctuations. The manuscript should state more carefully what qmin_rel can and cannot certify, or provide an additional characterization of the master-laser phase stability on the pulse-to-pulse timescale.
minor comments (5)
  1. [Sec. IIIA] The text says 'the ML will be refereed to as the local oscillator'; 'refereed' should be 'referred'.
  2. [Sec. IIA, references] In the sentence citing decoy-state security proofs, the reference marker '28?' appears with a stray question mark; this citation should be cleaned up.
  3. [Sec. IIIA; Eq. (7)] Equation (7) writes A_S(n)(τ_n) = (A_S(n)(τ_n))*, which is a tautology as written; the notation for the complex amplitude and its conjugate should be clarified.
  4. [Abstract; Sec. IIIE] The term 'source-agnostic' is used in the abstract and conclusions, but only a DFB slave laser is demonstrated; the method may be adaptable to other sources, but the claim of source-agnosticism should be softened unless a non-DFB device is tested.
  5. [General] No data availability statement or analysis code is provided; for a method whose quantitative output is intended for security certification, making the phase-extraction and fitting code available would substantially improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central isolation bound is an empirical measurement against the external control of injected optical power, not a quantity forced by the definition of qrel or by self-citation.

full rationale

The paper's central claim—that roughly 140 dB of isolation is needed against injection locking—is an experimentally derived output, not an input renamed as a result. The metric qrel is defined in Eq. (4) from the measured relative phase distribution and is measured via heterodyne detection (Eqs. (9)-(11)); the isolation figure follows from the empirically observed power threshold near -90 dBm in Fig. 7 combined with the external 100 W fiber damage threshold. No equation in the paper reduces the target quantity to its own definition. The wrapped Voigt fit in Sec. IIIC is explicitly labeled an assumption: 'We emphasize that the model chosen is part of the assumptions underlying the characterization', so the model is not covertly presented as first principles. The self-citations [8,31] are contextual and are not load-bearing for the isolation result. The paper repeatedly acknowledges the main limitation, that the characterization is only as good as the master laser's ability to simulate Eve's optimal attack (Secs. IIIA, IIIE, IV); this makes the quantitative bound conditional, but it is not a circular derivation because the paper does not claim to have proven optimality of the ML. The derivation chain is self-contained: the output is a measured value compared against external reference points, not a restatement of the metric or of a prior self-citation.

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

The central quantitative result depends on several unproven premises: the q-parameter security framework, the wrapped Voigt model used to extract qrel, the no-correlation baseline, the interferometer stability, and the 100 W damage threshold. The wrapped Voigt model and the master-laser equivalence to Eve are the most consequential; the paper flags both in the text.

free parameters (3)
  • Wrapped Voigt fit parameters (μ, σ, γ) = fitted per phase histogram via nonlinear regression
    The reported qrel(τn) is computed as 2π times the minimum of the fitted wrapped Voigt PDF (Eq. (12)-(14)), so the fitted parameters directly determine the central measured quantity. The authors acknowledge the model choice is part of the characterization assumptions.
  • Time window for phase extraction = centered window on the pulse, chosen so the model agrees with data (App. D, Fig. 8)
    The window selection affects which τn are included in qmin_rel = min_τn qrel(τn). Choosing the window on the basis of fit quality is a hand selection that can bias the reported minimum.
  • Injected power threshold for 'no significant influence' = approx. -90 dBm
    This threshold is read off the measured qmin_rel versus power curve (Fig. 7) and is the basis for the 140 dB isolation estimate; it is an empirically determined operating point, not a derived constant.
assumptions (5)
  • domain assumption Standard decoy-state BB84 security proofs require uniform or characterized phase randomization via the q-parameter framework
    The paper adopts the q-parameter from Currás-Lorenzo et al. and Nahar et al. to define qrel and to argue that characterizing qrel is sufficient for security evaluation (Sec. IIA, IIC). This is background theory the paper does not re-derive.
  • domain assumption Without injection locking, the slave laser's phases are independent and identically distributed with no cavity correlations
    Stated in Sec. IIC: 'we assume that no phase correlations are present in the laser cavity without an injection-locking attack.' The authors justify it by citing low correlations at 5 GHz and operating at 40 MHz, but it is a modeling assumption.
  • domain assumption Interferometer phase fluctuations are negligible over the 200 µs acquisition
    They measure '<5e-4 rad' drift by interfering the ML with itself via a Faraday mirror (Sec. IIIA). This is an experimentally supported assumption, not verified for all conditions.
  • ad hoc to paper The phase histogram is well described by a wrapped Voigt profile
    The model in Eq. (12) is chosen for the fit; the authors state 'the model chosen is part of the assumptions underlying the characterization' (Sec. IIIC). A different model would give different qmin_rel values.
  • domain assumption Eve's maximum injectable power is bounded by the fiber laser-induced damage threshold of 100 W
    Used in Sec. IIIE to convert the -90 dBm threshold into the 'at least about 140 dB' isolation requirement, citing Makarov et al. This is an external assumption from the certification literature.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Evaluation of quantum key distribution systems against injection-locking attacks." pith.science (2026). https://pith.science/paper/4TOTKQ76

@misc{pith2026241210290,
  author       = {Pith},
  title        = {Pith review of: Evaluation of quantum key distribution systems against injection-locking attacks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4TOTKQ76}},
  note         = {Machine review of arXiv:2412.10290}
}
read the original abstract

While ideal quantum key distribution (QKD) systems are well-understood, practical implementations face various vulnerabilities, such as side-channel attacks resulting from device imperfections. Current security proofs for decoy-state BB84 protocols either assume uniform phase randomization of Alice's signals, which is compromised by practical limitations and attacks like injection locking, or rely on a (partially) characterized phase distribution. This work presents an experimental method to characterize the phase de-randomization from injection locking using a heterodyne detection setup, providing a lower bound on the degree of isolation required to protect QKD transmitters against injection-locking attacks. The methods presented are source-agnostic and can be used to evaluate general QKD systems against injection-locking attacks.

Figures

Figures reproduced from arXiv: 2412.10290 by the authors.

Figure 1
Figure 1. Setup used to characterize the phase distribution under an injection-locking attack. The dashed boxes [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Spectra of the master laser (ML) and slave laser (SL) as well the SL spectrum under light injection for [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Exemplary oscilloscope homodyne detection [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Homodyne detection signals I (n) 0 (τn) and I (n) π/2 (τn), of the N = 8000 acquired slave laser pulses, plotted in phase space for τn = 6.56 ns with −42.3 dBm and without injection locking. C. Determining the degree of phase de-randomization from injection locking The…
Figure 5
Figure 5. Figure 5: Histogram of the phases for the N = 8000 slave laser pulses, determined from [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: q min rel -parameter versus polarization state of the light injected into the slave laser. The dashed (dotted) lines mark the polarization states with minimum (maximum) q min rel -parameter. The injected optical power was −55.3 dBm. a) Heatmap of the q min rel -paramet…
Figure 7
Figure 7. Figure 7: Degree of phase de-randomization, quantified by the [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 8
Figure 8. Figure 8: qrel-parameter and sum of squared residuals S 2 . The time window for the determination of the qrel-parameter is shaded in light blue. [1] C. H. Bennett and G. Brassard. Quantum cryptogra￾phy: Public key distribution and coin tossing. Theoret￾ical Computer Science, 560…

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. Beyond the Quantum Promise: A Security Analysis of Classical Control in Quantum Key Distribution

    quant-ph 2026-08 reject novelty 5.0 of 10

    Symbolic Tamarin models of standards-grounded QKD protocols reveal three control-plane vulnerabilities and confirm two countermeasures.

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages · cited by 1 Pith paper

  1. [1]

    C. H. Bennett and G. Brassard. Quantum cryptogra- phy: Public key distribution and coin tossing.Theoret- ical Computer Science, 560:7–11, 1984

    work page 1984

  2. [2]

    Gisin, G

    N. Gisin, G. Ribordy, W. Tittel, and H. Zbinden. Quan- tum cryptography.Rev. Mod. Phys., 74:145–195, 2002

    work page 2002

  3. [3]

    Scarani, H

    V. Scarani, H. Bechmann-Pasquinucci, N. J. Cerf, M. Dušek, N. Lütkenhaus, and M. Peev. The secu- rity of practical quantum key distribution.Rev. Mod. Phys., 81:1301–1350, 2009

    work page 2009

  4. [4]

    R. Renner. Security of Quantum Key Distribution, 2005

    work page 2005

  5. [5]

    Tomamichel, C

    M. Tomamichel, C. C. W. Lim, N. Gisin, and R. Ren- ner. Tight Finite-Key Analysis for Quantum Cryptog- raphy. Nature Communications, 3(1):634, 2012

    work page 2012

  6. [6]

    C. C. W. Lim, M. Curty, N. Walenta, F. Xu, and H. Zbinden. Concise security bounds for practical decoy-state quantum key distribution.Physical Review A, 89(2):022307, 2014. Publisher: American Physical Society

    work page 2014

  7. [7]

    Tomamichel and A

    M. Tomamichel and A. Leverrier. A largely self- 12 contained and complete security proof for quantum key distribution. Quantum, 1:14, 2017

    work page 2017

  8. [8]

    Rusca, A

    D. Rusca, A. Boaron, F. Grünenfelder, A. Mar- tin, and H. Zbinden. Finite-key analysis for the 1- decoy state QKD protocol. Applied Physics Letters, 112(17):171104, 2018

    work page 2018

Show all 32 references
  1. [9]

    Zapatero, Á

    V. Zapatero, Á. Navarrete, and M. Curty. Implementa- tion security in quantum key distribution, 2023

    2023

  2. [10]

    Currás-Lorenzo, M

    G. Currás-Lorenzo, M. Pereira, G. Kato, M. Curty, and K. Tamaki. A security framework for quantum key dis- tribution implementations, 2024

    2024

  3. [11]

    Sajeed, P

    S. Sajeed, P. Chaiwongkhot, A. Huang, H. Qin, V. Egorov, A. Kozubov, A. Gaidash, V. Chistiakov, A. Vasiliev, A. Gleim, and V. Makarov. An approach for security evaluation and certification of a complete quantum communication system. Scientific Reports, 11(1):5110, 2021

    2021

  4. [12]

    Makarov, A

    V. Makarov, A. Abrikosov, P. Chaiwongkhot, A. K. Fe- dorov, A. Huang, E. Kiktenko, M. Petrov, A. Ponosova, D. Ruzhitskaya, A. Tayduganov, D. Trefilov, and K. Za- itsev. Preparingacommercialquantumkeydistribution system for certification against implementation loop- holes, 2023

    2023

  5. [13]

    A Study on Implementation Attacks against QKD Systems

    BSI. A Study on Implementation Attacks against QKD Systems. Technical report, German Federal Office for Information Security, 2024

    2024

  6. [14]

    Tupkary, S

    D. Tupkary, S. Nahar, P. Sinha, and N. Lütkenhaus. Phase error rate estimation with basis-efficiency mis- match for decoy-state BB84, 2024

    2024

  7. [15]

    Sixto, Á

    X. Sixto, Á. Navarrete, M. Pereira, G. Currás-Lorenzo, K. Tamaki, and M. Curty. Quantum key distribution with imperfectly isolated devices, 2024

    2024

  8. [16]

    Gottesman, H.-K

    D. Gottesman, H.-K. Lo, N. Lütkenhaus, and J. Preskill. Security of quantum key distribution with imperfect devices, 2004

    2004

  9. [17]

    W.-Y. Hwang. Quantum Key Distribution with High Loss: Toward Global Secure Communication.Physical Review Letters, 91(5):057901, 2003

    2003

  10. [18]

    X.-B. Wang. Beating the photon-number-splitting at- tack in practical quantum cryptography.Physical Re- view Letters, 94(23):230503, 2005

    2005

  11. [19]

    H.-K. Lo, X. Ma, and K. Chen. Decoy State Quantum Key Distribution. Physical Review Letters, 94(23):230504, 2005

    2005

  12. [20]

    Sixto, G

    X. Sixto, G. Currás-Lorenzo, K. Tamaki, and M. Curty. Secret key rate bounds for quantum key distribution with faulty active phase randomization.EPJ Quantum Technology, 10(1), 2023

    2023

  13. [21]

    Nahar, T

    S. Nahar, T. Upadhyaya, and N. Lütkenhaus. Imper- fect Phase-Randomisation and Generalised Decoy-State Quantum Key Distribution. Physical Review Applied, 20(6):064031, 2023. arXiv:2304.09401 [quant-ph]

    2023 arXiv

  14. [22]

    Tang, H.-L

    Y.-L. Tang, H.-L. Yin, X. Ma, C.-H. F. Fung, Y. Liu, H.-L. Yong, T.-Y. Chen, C.-Z. Peng, Z.-B. Chen, and J.-W. Pan. Source attack of decoy-state quantum key distribution using phase information. Physical Review A, 88(2):022308, 2013

    2013

  15. [23]

    Pang, A.-L

    X.-L. Pang, A.-L. Yang, C.-N. Zhang, J.-P. Dou, H. Li, J. Gao, and X.-M. Jin. Hacking quantum key distribu- tion via injection locking.Phys. Rev. Appl., 13:034008, 2020

    2020

  16. [24]

    Huang, Á

    A. Huang, Á. Navarrete, S.-H. Sun, P. Chaiwongkhot, M. Curty, and V. Makarov. Laser-Seeding Attack in Quantum Key Distribution. Physical Review Applied, 12(6):064043, 2019

    2019

  17. [25]

    Zhang, M.-S

    X.-X. Zhang, M.-S. Jiang, Y. Wang, Y.-F. Lu, H.-W. Li, C. Zhou, Y. Zhou, and W.-S. Bao. Analysis of an injection-locking-loophole attack from an external source for quantum key distribution. Phys. Rev. A, 106:062412, 2022

    2022

  18. [26]

    Lovic, D.G

    V. Lovic, D.G. Marangon, P.R. Smith, R.I. Woodward, and A.J. Shields. Quantified effects of the laser-seeding attack in quantum key distribution. Physical Review Applied, 20(4):044005, 2023

    2023

  19. [27]

    Currás-Lorenzo, S

    G. Currás-Lorenzo, S. Nahar, N. Lütkenhaus, K. Tamaki, and M. Curty. Security of quantum key distribution with imperfect phase randomisation. Quantum Science and Technology, 9(1):015025, 2024. arXiv:2210.08183 [quant-ph]

    2024 arXiv

  20. [28]

    Lo and J

    H.-K. Lo and J. Preskill. Security of quantum key dis- tribution using weak coherent states with nonrandom phases. 2007

    2007

  21. [29]

    R. J. Glauber. Coherent and incoherent states of the radiation field. Phys. Rev., 131:2766–2788, 1963

    1963

  22. [30]

    Z. Cao, Z. Zhang, H.-K. Lo, and X. Ma. Discrete-phase- randomized coherent state source and its application in quantum key distribution.New Journal of Physics, 17(5):053014, 2015

    2015

  23. [31]

    Grünenfelder, A

    F. Grünenfelder, A. Boaron, D. Rusca, A. Martin, and H. Zbinden. Performance and security of 5 GHz repeti- tion rate polarization-based quantum key distribution. Applied Physics Letters, 117(14):144003, 2020

    2020

  24. [32]

    S. Huet, A. Bouvier, M.-A. Poursat, and E. Jolivet. Statistical Tools for Nonlinear Regression A Practical Guide With S-PLUS and R Examples. Springer-Verlag, 2004

    2004

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.