Pith. sign in

REVIEW 2 major objections 5 minor 70 references

Heisenberg-limited continuous-variable distributed quantum metrology with arbitrary weights

T0 review · 2 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Two light inputs, one quantum and one classical, are necessary and sufficient for Heisenberg-limited estimation of arbitrary weighted sums of distributed phases, with a tight bound set by the nonclassical state's metrological power.

desk verdict Solid minimality theorem and explicit construction for CV distributed sensing, but the claimed 'universal' lower bound rests on an approximate inverse that the authors themselves flag. read the letter →

arxiv 2412.01074 v2 pith:T5B5CTFJ submitted 2024-12-02 quant-ph

Wenchao Ge , Kurt Jacobs This is my paper · ORCID
classification quant-ph
keywords distributedquantummetrologycontinuousvariablesHeisenberglimitFisherinformationmetrologicalpowersqueezedvacuumcatstateslinearopticalnetworks
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

Distributed quantum metrology asks how many resources a network needs to measure many unknown phases with the best precision quantum mechanics allows. This paper shows that two non-vacuum inputs — one nonclassical state (for example squeezed vacuum or a cat state) and one ordinary coherent laser — are both necessary and sufficient to estimate any balanced weighted sum of $d$ distributed phases at the Heisenberg limit, where the error scales as $1/N$ with $N$ the total photon number. The central result is a universal error bound $\Delta q \ge 1/\sqrt{N + 2 n_2 W}$, in which $n_2$ is the coherent input's mean photon number and $W$ is the metrological power of the nonclassical state; the bound is tight for squeezed vacuum, cat, squeezed-thermal, and Fock states. A single non-vacuum input cannot do the job for weights of both signs, because it produces only positive-weight sensitivity. The paper also shows that the same network estimates arbitrary analytic functions of the phases, and that local photon-number detection saturates the bound for a useful class of states.

What carries the argument

The machinery is the rank-plus-diagonal decomposition of the QFIM, $F = c_u uu^T + c_v vv^T + c_s(uv^T + vu^T) + N$, with $u_j = |U_{j1}|^2$ and $v_j = U_{j1} U_{j2}^*$ determined by the first two columns of the network unitary. The scalar $W$ is the metrological power of the nonclassical state, its maximal quantum advantage for quadrature sensing, given for pure states by $W = n_1 - |\alpha_1|^2 + |\xi_1 - \alpha_1^2|$; this quantity controls the quantum-enhanced term $c_v = 8 n_2 W$. The argument proceeds by forcing the estimated weight vector $w$ to be parallel to $v$, which is the only zero-sum network vector, and then inverting the QFIM to obtain the bound. The same decomposition also identifies which input states help: only the first three moments of the annihilation operator enter $W$ and $c_s$, so a wide class of nonclassical states, not just squeezed light, can supply the quantum advantage.

What would settle it

Numerically optimize the full, unapproximated quantum Fisher information matrix for a small network (for instance $d=2$) over passive unitaries $U$ and over the photon-number split between a squeezed-vacuum input and a coherent input at fixed total $N$, computing the exact inverse including the $N$ term; any configuration with $\Delta q < 1/\sqrt{N + 2 n_2 W}$ would disprove the universal bound. A complementary check is to test whether a single non-vacuum input with a weight vector containing both signs can beat the SQL, which the paper says is impossible.

Watch

Extended reading notes

Core claim

The paper's central discovery is the complete structure of the quantum Fisher information matrix (QFIM) for any passive linear network fed by one nonclassical state plus one coherent state: $F = c_u u u^T + c_v v v^T + c_s (u v^T + v u^T) + N$, where $u$ and $v$ are determined by the network's first two columns, $c_v = 8 n_2 W$ is set by the coherent state's energy times the nonclassical state's metrological power, and $N$ is the standard-quantum-limit diagonal term. Because the phase-reference condition forces every valid weight vector $w$ to have zero sum, and because $v$ is the only network vector with zero sum, the high-energy part of the estimation must come from $v$; the explicit choice $U_{j1}=\sqrt{|w_j|}$, $U_{j2}=w_j/\sqrt{|w_j|}$ makes $v=w$ and gives $\Delta q \ge 1/\sqrt{N + 2 n_2 W}$. The bound is saturated whenever the cross term $c_s$ vanishes, which includes squeezed vacuum, cat, squeezed-thermal, and Fock states. This establishes necessity and sufficiency of two inputs: with only one non-vacuum input the QFIM reduces to $c_u u u^T + N$, and $u$ has strictly positive entries, so no Heisenberg-limited estimation of a linear combination containing both positive and negative weights is possible.

Load-bearing premise

The universal part of the bound depends on the supplement's approximate inversion of the Fisher matrix, where the standard-quantum-limit diagonal term is expanded with off-diagonal pieces neglected and the network condition $u \perp v$ is known to be optimal only when that term is absent; if some network exploits the neglected term to do better, the claimed universality could fail, although the explicit construction still reaches the Heisenberg limit.

Editorial extensions

If this is right

  • Two input ports are minimal: no single non-vacuum input can provide Heisenberg-limited estimation of an arbitrary linear combination with both positive and negative weights, while one nonclassical plus one coherent input can.
  • The universal bound is tight for every nonclassical state with vanishing cross term $c_s$, so squeezed vacuum, cat, squeezed-thermal, and Fock states saturate it with a suitable passive network.
  • In the weak-nonclassical regime ($1 < W \ll n_2$), the network still multiplies the classical sensitivity by about $\sqrt{1 + 2W}$, so modest squeezing can enhance sensing with strong lasers.
  • The same network estimates arbitrary analytic functions $f(\theta)$ with variance $\|\nabla f\|_1^2/(4N + 8 n_2 W)$, reaching Heisenberg scaling under the optimal two-stage resource split.
  • For input states satisfying the moment conditions $\alpha_1 = \beta_1 = 0$ plus a reality condition, photon-number-resolving detection at each node saturates the quantum bound.

Reading between the lines

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

  • Read as a recipe, the paper implies that the hard resource to supply is metrological power $W$ rather than high photon number in the nonclassical state; a modest squeezed source backed by a strong coherent beam is the practical configuration.
  • Because $W$ is defined for quadrature sensing and is independent of the network, the same two-input minimality likely transfers to distributed displacement sensing, though the paper only analyzes phase shifts.
  • A natural next test is whether a compensating output unitary $V$ can make local photon-number detection saturate the bound for states with nonzero $c_s$; the paper's saturation proof assumes $c_s = 0$.
  • The QFIM decomposition suggests a resource-theoretic reading: among all nonclassical states at fixed energy, the one maximizing $W$ (squeezed vacuum) should also be optimal for arbitrary distributed weights, generalizing the two-mode result.
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

2 major / 5 minor

Summary. The paper studies continuous-variable distributed quantum metrology in a linear optical network with two non-vacuum inputs: one single-mode nonclassical state (pure or mixed) and one coherent state, with 2d modes arranged as d phase-reference pairs. The central object is the quantum Fisher information matrix, which the authors decompose as F = c_u uu^T + c_v vv^T + c_s(uv^T+vu^T) + N, Eq. (3), where u and v encode the network action and c_v = 8 n_2 W with W the metrological power. On this basis the paper claims: (i) two inputs, only one nonclassical, are necessary and sufficient for Heisenberg-limited estimation of an arbitrary linear combination of the d phases; (ii) a universal lower bound of the form Δq ≥ 1/√(N + 2 n_2 W), Eq. (4), tight when c_s = 0; (iii) local photon-number detection saturates this bound for a class of nonclassical states; (iv) arbitrary analytic functions of the phases can be estimated with the same scaling; and (v) the network has two operating regimes, one at the Heisenberg limit and one where a weak nonclassical state multiplicatively enhances classical sensitivity. The supplemental material contains the inverse-QFIM calculations, the mixed-state generalization, a (d+1)-mode reduced scheme, a four-mode example, and the photon-counting optimality derivation.

Significance. If the universal bound is valid, this is a substantial and useful unification: it reduces to known two-mode Mach-Zehnder results at d = 1, identifies the metrological power W as the resource controlling Heisenberg scaling, and provides a concrete four-mode network that realizes arbitrary two-parameter weights. The constructive part is carefully executed and convincing: for the choice U_j1 = √|w_j| and U_j2 = w_j/√|w_j| one has v = w, u ⟂ v, and N = 4N diag(|w_j|), so that the inverse in Eq. (S.14) is exact and Eq. (S.15) gives the advertised sensitivity; the d=1 limit checks out; and the extension to mixed input states through generalized coefficients in Supplemental S4 is well organized. The gap is in the converse direction: the claim that no other linear network can do better than the constructed one is not established, because the optimization over the network in Supplemental S2 is performed with an approximate inverse. The paper is therefore a strong candidate for publication once the universality claim is either proven exactly or properly restricted.

major comments (2)
  1. [Eq. (4) and Supplemental S2] The claimed universal lower bound is not proven as stated. The derivation of the optimal alignment uses Eq. (S.12), in which the inverse of F is approximated by dropping the off-diagonal terms e_j^T N n_j, and the condition u ⟂ v is carried over from the N = 0 analysis. The authors explicitly acknowledge in S2 that "we neglected the off-diagonal terms when we expanded N" and that "v⊥u may not always hold" when N is included. Since Eq. (4) is used to conclude that no network can beat the constructed one, this is a load-bearing step. I ask the authors either to provide an exact proof that for every unitary U the quantity w^T F^{-1}w is bounded below by 1/(4N + c_v - c_s^2/(4N + c_u)), or to restate Eq. (4) as the sensitivity of the explicitly constructed aligned network and soften the "universal" and "minimum" claims accordingly.
  2. [Single-mode input paragraph] The necessity of two inputs is argued from the N = 0 form F ≈ c_u uu^T: since u has positive entries, a signed weight vector cannot be parallel to u, so no Heisenberg scaling is obtained. This is a heuristic statement: once N is included, the optimal weighting vector is not exactly parallel to u, and the proof in S2 that the N-induced component contributes only at order 1/O(N) rests on the same approximate inverse flagged in the previous comment. To make the claim that two inputs are the minimum required, the single-input case needs a direct bound showing that for every network, w^T(c_u uu^T + N)^{-1}w cannot scale as 1/N^2 when w has both positive and negative entries.
minor comments (5)
  1. [Abstract and Introduction] The phrase "universal and tight upper bound on the sensitivity" is misleading: Eq. (4) is a lower bound on the estimation error Δq. Please reword as a lower bound on uncertainty or equivalently an upper bound on sensitivity only if sensitivity is defined as the inverse error.
  2. [Model section] There are several typographical errors: "discucssed" should be "discussed", and "su fficient" should be "sufficient". Please proofread the LaTeX source.
  3. [Supplemental S2, Eq. (S.13)] The notation 1/O(N) in Eq. (S.13) is informal: it should say that the term is of order 1/N with the constant depending on the omitted cross terms, rather than using the symbol O(N) as a scalar denominator.
  4. [Supplemental S6, Eq. (S.26)] The saturation result for local photon-number detection matches Eq. (4) only when α1 = β1 = 0 and either the weights are uniform, (||w||_3)^3 = |w|^4, or n1, n2 ≫ 1. The main text should state these conditions explicitly when claiming that photon-number detection achieves the maximum sensitivity, rather than leaving them only in the supplement.
  5. [Figure 2] The lower panel legend would benefit from stating explicitly that the solid curves are from Eq. (4) and that the dashed lines mark the SQL and HL scaling; currently the distinction between solid and dashed curves is not described in the caption.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the bound is a direct QFIM calculation; the acknowledged S2 approximations are a rigor gap, not a circular reduction.

full rationale

The paper's central derivation is self-contained. Equation (3) expresses the QFIM directly in terms of the input state moments and the network entries U_jk, and the bound in Eq. (4) is obtained by inverting that QFIM. The metrological power W appears in Eq. (3) as an explicit function of state moments (W = n1 - |alpha1|^2 + |xi1 - alpha1^2|); the citations [45,46] attach the name and known extremal properties to this already-defined quantity, but they are not used to substitute the target inequality into itself. The 'tightness' part of the claim is constructive: the authors exhibit U_j1 = sqrt(|w_j|), U_j2 = w_j/sqrt(|w_j|), giving v = w and u perpendicular to v, and Eq. (S.14) is then an exact inverse under that alignment. Thus the achievability of 1/sqrt(N + 2n2W) follows from the stated model, not from a fitted parameter or a renamed input. The function-estimation result, Eq. (6), is inherited from the two-step protocol of Qian et al. [16] combined with the derived sensitivity, and does not smuggle the conclusion into the premise. The only genuine weakness is in Supplemental S2, where the authors explicitly write: 'We note that the treatment of N has two limitations. First, we neglected the off-diagonal terms when we expanded N in the orthonormal basis of u and v. Second, the condition v perpendicular to u was optimal without the term N. In optimizing the full QFIM, v perpendicular to u may not always hold.' This is an honest proof gap: the universal part of Eq. (4) is not fully established for all networks because the inverse used to minimize over networks is approximate. But a proof gap is not circularity. No equation in the derivation reduces by construction to its own input, no fitted quantity is relabeled as a prediction, and no load-bearing uniqueness theorem is imported from the authors' prior work to forbid alternatives. The paper is therefore scored 0 for circularity, with the S2 limitation flagged as a correctness risk rather than a circular step.

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

The central results rest on the standard formalism of quantum Fisher information, on the assumed product input structure of one nonclassical state and one coherent state, on the phase-reference constraint Σ w_j = 0, and on imported definitions of metrological power from prior work by the same group. No invented entities or data-fitted parameters are introduced.

assumptions (5)
  • domain assumption The initial state is a product of one single-mode nonclassical state, one coherent state, and vacuum in all other modes.
    Stated in 'The Model': |Ψ⟩ = |ψ⟩⊗|α⟩⊗|0...⟩; the entire QFIM calculation rests on this input structure.
  • domain assumption Each measured phase is paired with a local reference phase, giving 2d modes and weight vectors satisfying Σ_j w_j = 0.
    Stated after Eq. (1): w_{2i-1} = -w_{2i}; this restricts the class of estimable global functions to phase differences, which is physically required for phase estimation.
  • domain assumption The network is a passive linear-optical unitary U acting on 2d modes.
    The model in 'The Model' assumes U is a linear optical network; all results apply only to such networks.
  • domain assumption Metrological power W as defined in refs [45,46] has the stated properties and bounds.
    The bound in Eq. (4) is expressed via W; the paper imports W's definition and scaling from prior work by the same authors, which is legitimate but is an input from outside this paper.
  • ad hoc to paper For the universal lower bound, the standard quantum limit matrix N can be treated in the inverse as in S2, neglecting certain off-diagonal terms.
    S2 states: treatment of N has two limitations, neglected off-diagonal terms and possible non-orthogonality of u and v; the universal claim in Eq. (4) rests on this approximate treatment.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Heisenberg-limited continuous-variable distributed quantum metrology with arbitrary weights." pith.science (2026). https://pith.science/paper/T5B5CTFJ

@misc{pith2026241201074,
  author       = {Pith},
  title        = {Pith review of: Heisenberg-limited continuous-variable distributed quantum metrology with arbitrary weights},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T5B5CTFJ}},
  note         = {Machine review of arXiv:2412.01074}
}
read the original abstract

Distributed quantum metrology (DQM) enables the estimation of global functions of d distributed parameters beyond the capability of separable sensors. Continuous-variable DQM involves using a linear network with at least one nonclassical input. Here we fully elucidate the structure of linear networks with two non-vacuum inputs which allows us to prove a number of fundamental properties of continuous-variable DQM. While measuring the sum of d parameters at the Heisenberg limit can be achieved with a single non-vacuum input, we show that two inputs, one of which can be classical, is required to measure an arbitrary linear combination of d parameters and an arbitrary global function of the parameters. We obtain a universal and tight upper bound on the sensitivity of DQM networks with two inputs, and completely characterize the properties of the nonclassical input required to obtain a quantum advantage. This reveals that a wide range of nonclassical states make this possible, including a squeezed vacuum. We also show that for a class of nonclassical inputs local photon number detection will achieve the maximum sensitivity. Finally we show that a general DQM network has two distinct regimes. The first achieves Heisenberg scaling. In the second the nonclassical input is much weaker than the coherent input, nevertheless providing a multiplicative enhancement to the otherwise classical sensitivity.

Figures

Figures reproduced from arXiv: 2412.01074 by the authors.

Figure 1
Figure 1. FIG. 1. Continuous-variable distributed quantum metrology proto [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. (Color online). (a) The sensitivity scaling, [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. A four-mode linear network [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

70 extracted references · 55 canonical work pages

  1. [1]

    Giovannetti, S

    V . Giovannetti, S. Lloyd, and L. Maccone, Phys. Rev. Lett. 96, 010401 (2006)

    2006

  2. [2]

    P. C. Humphreys, M. Barbieri, A. Datta, and I. A. Walmsley, Physical Review Letters 111, 070403 (2013)

    work page 2013

  3. [3]

    C. N. Gagatsos, D. Branford, and A. Datta, Phys. Rev. A 94, 042342 (2016)

    work page 2016

  4. [4]

    Polino, M

    E. Polino, M. Riva, M. Valeri, R. Silvestri, G. Corrielli, A. Crespi, N. Spagnolo, R. Osellame, and F. Sciarrino, Optica 6, 288 (2019)

    work page 2019

  5. [5]

    Szczykulska, T

    M. Szczykulska, T. Baumgratz, and A. Datta, Advances in Physics: X 1, 621 (2016)

    2016

  6. [6]

    S. Z. Ang, R. Nair, and M. Tsang, Phys. Rev. A 95, 063847 (2017)

    work page 2017

  7. [7]

    Albarelli, M

    F. Albarelli, M. Barbieri, M. Genoni, and I. Gianani, Physics Letters A 384, 126311 (2020)

    2020

  8. [8]

    Eldredge, M

    Z. Eldredge, M. Foss-Feig, J. A. Gross, S. L. Rolston, and A. V . Gorshkov, Phys. Rev. A97, 042337 (2018)

    work page 2018

Show all 70 references
  1. [9]

    W. Ge, K. Jacobs, Z. Eldredge, A. V . Gorshkov, and M. Foss- Feig, Phys. Rev. Lett. 121, 043604 (2018)

    2018

  2. [10]

    Zhuang, Z

    Q. Zhuang, Z. Zhang, and J. H. Shapiro, Phys. Rev. A 97, 032329 (2018)

    2018

  3. [11]

    T. J. Proctor, P. A. Knott, and J. A. Dunningham, Phys. Rev. Lett. 120, 080501 (2018)

    2018

  4. [12]

    Komar, E

    P. Komar, E. M. Kessler, M. Bishof, L. Jiang, A. S. Sørensen, J. Ye, and M. D. Lukin, Nature Physics 10, 582 (2014)

    2014

  5. [13]

    Majumder, L

    S. Majumder, L. Andreta de Castro, and K. R. Brown, npj Quantum Information 6, 19 (2020)

    2020

  6. [14]

    Y . Xia, W. Li, W. Clark, D. Hart, Q. Zhuang, and Z. Zhang, Phys. Rev. Lett. 124, 150502 (2020)

    2020

  7. [15]

    X. Sun, W. Li, Y . Tian, F. Li, L. Tian, Y . Wang, and Y . Zheng, Photon. Res. 10, 2886 (2022)

    2022

  8. [16]

    K. Qian, Z. Eldredge, W. Ge, G. Pagano, C. Monroe, J. V . Porto, and A. V . Gorshkov, Phys. Rev. A100, 042304 (2019)

    2019

  9. [17]

    Rubio, P

    J. Rubio, P. A. Knott, T. J. Proctor, and J. A. Dunningham, Journal of Physics A: Mathematical and Theoretical53, 344001 (2020)

    2020

  10. [18]

    T. Qian, J. Bringewatt, I. Boettcher, P. Bienias, and A. V . Gor- shkov, Phys. Rev. A103, L030601 (2021)

    2021

  11. [19]

    Bringewatt, I

    J. Bringewatt, I. Boettcher, P. Niroula, P. Bienias, and A. V . Gorshkov, Phys. Rev. Res.3, 033011 (2021)

    2021

  12. [20]

    Ehrenberg, J

    A. Ehrenberg, J. Bringewatt, and A. V . Gorshkov, Phys. Rev. Res. 5, 033228 (2023)

    2023

  13. [21]

    Giovannetti, S

    V . Giovannetti, S. Lloyd, and L. Maccone, Science 306, 1330 (2004)

    2004

  14. [22]

    Zwierz, C

    M. Zwierz, C. A. P ´erez-Delgado, and P. Kok, Physical review letters 105, 180402 (2010)

    2010

  15. [23]

    Y . Yang, B. Yadin, and Z.-P. Xu, Phys. Rev. Lett.132, 210801 (2024)

    2024

  16. [24]

    Tiedau, T

    J. Tiedau, T. J. Bartley, G. Harder, A. E. Lita, S. W. Nam, T. Gerrits, and C. Silberhorn, Phys. Rev. A100, 041802 (2019)

    2019

  17. [25]

    M. Uria, P. Solano, and C. Hermann-Avigliano, Phys. Rev. Lett. 125, 093603 (2020)

    2020

  18. [26]

    Rivera, J

    N. Rivera, J. Sloan, Y . Salamin, J. D. Joannopoulos, and M. Soljacic, Proceedings of the National Academy of Sciences 120, e2219208120 (2023)

    2023

  19. [27]

    X. Deng, S. Li, Z.-J. Chen, Z. Ni, Y . Cai, J. Mai, L. Zhang, P. Zheng, H. Yu, C.-L. Zou, et al. , arXiv preprint arXiv:2306.16919 (2023)

    2023

  20. [28]

    B. Bell, S. Kannan, A. McMillan, A. S. Clark, W. J. Wadsworth, and J. G. Rarity, Phys. Rev. Lett.111, 093603 (2013)

    2013

  21. [29]

    S. Hong, J. ur Rehman, Y .-S. Kim, Y .-W. Cho, S.-W. Lee, H. Jung, S. Moon, S.-W. Han, and H.-T. Lim, Nature Com- munications 12, 5211 (2021)

    2021

  22. [30]

    Liu, Y .-Z

    L.-Z. Liu, Y .-Z. Zhang, Z.-D. Li, R. Zhang, X.-F. Yin, Y .-Y . Fei, L. Li, N.-L. Liu, F. Xu, Y .-A. Chen, and J.-W. Pan, Nature Photonics 15, 137 (2021)

    2021

  23. [31]

    Zhao, Y .-Z

    S.-R. Zhao, Y .-Z. Zhang, W.-Z. Liu, J.-Y . Guan, W. Zhang, C.- L. Li, B. Bai, M.-H. Li, Y . Liu, L. You, J. Zhang, J. Fan, F. Xu, Q. Zhang, and J.-W. Pan, Phys. Rev. X 11, 031009 (2021)

    2021

  24. [32]

    Valeri, V

    M. Valeri, V . Cimini, S. Piacentini, F. Ceccarelli, E. Polino, F. Hoch, G. Bizzarri, G. Corrielli, N. Spagnolo, R. Osellame, and F. Sciarrino, Phys. Rev. Res.5, 013138 (2023)

    2023

  25. [33]

    D.-H. Kim, S. Hong, Y .-S. Kim, Y . Kim, S.-W. Lee, R. C. Pooser, K. Oh, S.-Y . Lee, C. Lee, and H.-T. Lim, Nature com- munications 15, 266 (2024)

    2024

  26. [34]

    S. L. Braunstein and P. van Loock, Rev. Mod. Phys. 77, 513 (2005)

    2005

  27. [35]

    Gatto, P

    D. Gatto, P. Facchi, F. A. Narducci, and V . Tamma, Phys. Rev. Res. 1, 032024 (2019)

    2019

  28. [36]

    X. Guo, C. R. Breum, J. Borregaard, S. Izumi, M. V . Larsen, T. Gehring, M. Christandl, J. S. Neergaard-Nielsen, and U. L. Andersen, Nature Physics 16, 281 (2020)

    2020

  29. [37]

    van Loock and S

    P. van Loock and S. L. Braunstein, Phys. Rev. Lett. 84, 3482 (2000)

    2000

  30. [38]

    Yukawa, R

    M. Yukawa, R. Ukai, P. van Loock, and A. Furusawa, Phys. Rev. A 78, 012301 (2008)

    2008

  31. [39]

    V ogel and J

    W. V ogel and J. Sperling, Phys. Rev. A89, 052302 (2014)

    2014

  32. [40]

    Zhong, H

    H.-S. Zhong, H. Wang, Y .-H. Deng, M.-C. Chen, L.- C. Peng, Y .-H. Luo, J. Qin, D. Wu, X. Ding, Y . Hu, P. Hu, X.-Y . Yang, W.-J. Zhang, H. Li, Y . Li, X. Jiang, L. Gan, G. Yang, L. You, Z. Wang, L. Li, N.-L. Liu, 6 C.-Y . Lu, and J.-W. Pan, Science 370, 1460 (2020), https://w...

    2020 doi

  33. [41]

    L. S. Madsen, F. Laudenbach, M. F. Askarani, F. Rortais, T. Vincent, J. F. F. Bulmer, F. M. Miatto, L. Neuhaus, L. G. Helt, M. J. Collins, A. E. Lita, T. Gerrits, S. W. Nam, V . D. Vaidya, M. Menotti, I. Dhand, Z. Vernon, N. Quesada, and J. Lavoie, Nature 606, 75 (2022)

    2022

  34. [42]

    C. Oh, C. Lee, S. H. Lie, and H. Jeong, Phys. Rev. Res. 2, 023030 (2020)

    2020

  35. [43]

    Malitesta, A

    M. Malitesta, A. Smerzi, and L. Pezz `e, Phys. Rev. A 108, 032621 (2023)

    2023

  36. [44]

    Since the classical input does not distribute entanglement it can either be distributed locally from a single input, simpler from the point of view of maintaining a single phase reference, or be replaced by separate classical inputs that are local to the param- eters being measured

  37. [45]

    W. Ge, K. Jacobs, S. Asiri, M. Foss-Feig, and M. S. Zubairy, Phys. Rev. Res. 2, 023400 (2020)

    2020

  38. [46]

    W. Ge, K. Jacobs, and M. S. Zubairy, npj Quantum Information 9, 5 (2023)

    2023

  39. [47]

    Ge and M

    W. Ge and M. S. Zubairy, Phys. Rev. A 102, 043703 (2020)

    2020

  40. [48]

    Ourjoumtsev, H

    A. Ourjoumtsev, H. Jeong, R. Tualle-Brouri, and P. Grangier, Nature 448, 784 (2007)

    2007

  41. [49]

    Lewenstein, M

    M. Lewenstein, M. F. Ciappina, E. Pisanty, J. Rivera-Dean, P. Stammer, T. Lamprou, and P. Tzallas, Nature Physics 17, 1104 (2021)

    2021

  42. [50]

    See the supplementary materials for detailed derivations on the inverse of the QFIM, (d+1)−mode reduced scheme, generic in- put state, and the optimal measurement scheme, which includes Refs. [68–70]

  43. [51]

    Jarzyna and R

    M. Jarzyna and R. Demkowicz-Dobrza ´nski, Phys. Rev. A 85, 011801 (2012)

    2012

  44. [52]

    An equivalent scheme of d + 1 modes with only one reference mode is shown in the Supplemental Materials [50]

  45. [53]

    J. Liu, H. Yuan, X.-M. Lu, and X. Wang, Journal of Physics A: Mathematical and Theoretical 53, 023001 (2019)

    2019

  46. [54]

    S. M. Kay, Fundamentals of statistical signal processing (Pren- tice Hall PTR, 1993)

    1993

  47. [55]

    Takeoka, K

    M. Takeoka, K. P. Seshadreesan, C. You, S. Izumi, and J. P. Dowling, Phys. Rev. A 96, 052118 (2017)

    2017

  48. [56]

    Pezz ´e and A

    L. Pezz ´e and A. Smerzi, Phys. Rev. Lett. 100, 073601 (2008)

    2008

  49. [57]

    M. D. Lang and C. M. Caves, Phys. Rev. Lett. 111, 173601 (2013)

    2013

  50. [58]

    Meylahn, B

    F. Meylahn, B. Willke, and H. Vahlbruch, Phys. Rev. Lett.129, 121103 (2022)

    2022

  51. [59]

    Achilles, C

    D. Achilles, C. Silberhorn, S. Cezary, K. Banaszek, I. A. Walmsley, M. J. Fitch, B. C. Jacobs, T. B. Pittman, and J. D. Franson, Journal of Modern Optics 51, 1499 (2004), https://doi.org/10.1080/09500340408235288

    2004 doi

  52. [60]

    Kardynał, Z

    B. Kardynał, Z. Yuan, and A. Shields, Nature Photonics 2, 425 (2008)

    2008

  53. [61]

    Hlou ˇsek, M

    J. Hlou ˇsek, M. Dudka, I. Straka, and M. Jeˇzek, Phys. Rev. Lett. 123, 153604 (2019)

    2019

  54. [62]

    R. J. Birrittella, P. M. Alsing, and C. C. Gerry, A VS Quantum Science 3, 014701 (2021), https: //pubs.aip.org/avs/aqs/article- pdf/doi/10.1116/5.0026148/19739025/014701 1 online.pdf

    2021 doi

  55. [63]

    Oh, S.-Y

    C. Oh, S.-Y . Lee, H. Nha, and H. Jeong, Phys. Rev. A 96, 062304 (2017)

    2017

  56. [64]

    J. Yu, Y . Qin, J. Qin, H. Wang, Z. Yan, X. Jia, and K. Peng, Phys. Rev. Appl. 13, 024037 (2020)

    2020

  57. [65]

    T. Park, H. Stokowski, V . Ansari, S. Gyger, K. K. S. Multani, O. T. Celik, A. Y . Hwang, D. J. Dean, F. Mayor, T. P. McKenna, M. M. Fejer, and A. Safavi- Naeini, Science Advances 10, eadl1814 (2024), https://www.science.org/doi/pdf/10.1126/sciadv.adl1814

    2024 doi

  58. [66]

    Hou, J.-D

    L.-L. Hou, J.-D. Zhang, K.-M. Zheng, and S. Wang, Optics Communications 570, 130898 (2024)

    2024

  59. [67]

    Pezz `e and A

    L. Pezz `e and A. Smerzi, arXiv preprint arXiv:2405.18404 (2024)

    2024 arXiv

  60. [68]

    K. S. Miller, Mathematics Magazine 54, 67 (1981), https://doi.org/10.1080/0025570X.1981.11976898

    1981

  61. [69]

    H. Kwon, K. C. Tan, T. V olkoff, and H. Jeong, Phys. Rev. Lett. 122, 040503 (2019)

    2019

  62. [70]

    Tan, J.-Q

    Q.-S. Tan, J.-Q. Liao, X. Wang, and F. Nori, Phys. Rev. A 89, 053822 (2014). 7 Supplemental Materials S1. The lower bound on wTF−1w withoutN S1.1. Decomposition of the weighting vector In the limit of quantum-enhanced metrology, we neglect N in Eq. (3) in the main text. For Di...

    2014

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.