Multiparameter quantum bounds for entanglement-assisted aperture synthesis
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.
Kai: We're looking at the paper: "Multiparameter quantum bounds for entanglement-assisted aperture synthesis".
Title and authors: Kai: So we're looking at this paper 'Multiparameter quantum bounds for entanglement-assisted aperture synthesis' by Kalaga Madhav, Pieter Kok, and Nic Scott. It tackles the idea that we can get better imaging than just looking at a single baseline visibility when you have an array of stations.
Mira: Right, it gets pretty interesting because the authors are taking something that usually only studies one baseline and expanding it to an M-station array looking at an extended scene. They're framing this as a multiparameter estimation problem for the whole thing.
Lev: And the core idea they lay out is that they reduce this multimode optical problem down to the quantum Fisher information of a single delocalized photon, which is where things get mathematically tricky, I guess.
Kai: Exactly. They use a reduction lemma to show that this multimode quantum Fisher information equals the mean photon number times the QFI of one delocalized photon > That sounds like they're trying to simplify the complexity down to something we can handle with a single state.
Mira: But then they immediately run into trouble when they look at the computation of that QFI matrix and its mean Uhlmann curvature, and numerically it shows that for a point source, it is measurement-compatible, but for every extended source model they checked, it turns out to be incompatible > That's a big hurdle.
Lev: Incompatibility means there's a measurable gap between what the quantum mechanics allow and what we can actually achieve with measurements > So they are saying that no single measurement optimized parameter by parameter can reach the absolute best possible result.
Kai: And they quantify this incompatibility by showing the Holevo bound exceeds the symmetric logarithmic derivative bound by up to about seventy-four percent > That's a lot of inefficiency just from the measurement setup itself.
Mira: It means that even if we use ideal resources, we can only get a fraction of what's theoretically possible for all parameters at once > It sets a hard limit on what this kind of imaging technique can actually deliver right now.
Lev: Before we get into how they try to beat these limits, I want to mention their comparison with receivers. They explicitly compare an explicit collective receiver against a global mode-sorting measurement and a pairwise beam-splitter receiver > This is important because it shows that the way you collect the data matters a lot.
Kai: And what's really compelling is the result there: for two points on a six station array, this collective receiver gets within about seven percent of the symmetric logarithmic derivative bound > That's significantly better than whatever those pairwise receivers are doing.
Mira: It’s an order of magnitude worse, actually > So they are demonstrating that a collective receiver is way outperforming processing just the individual baseline visibilities.
Lev: This leads into their resource allocation section, where they discuss the noisy-resource advantage threshold > This threshold tells us when the whole array protocol actually becomes useful in practice.
Kai: They find a break-even baseline of about twenty kilometers for near-term parameters > So that sets a rough distance scale for what's feasible with current technology.
Mira: And they find that the optimal way to spend your entanglement budget is a convex program that gives them this closed form square root law solution > This rule suggests you should concentrate the entanglement on the lowest visibility baselines first.
Lev: So, if you have limited resources, this is how you should allocate them for best performance in terms of maximizing that advantage > It’s a practical prescription for building these kinds of arrays.
Kai: The conclusion says that quantum-enhanced imaging precision scales linearly with the number of photons collected and it's governed by a single M level state > That's a neat scaling law to have, suggesting linear improvement with more light.
Mira: But they also point out something important: these are ultimate quantum bounds, assuming ideal resources, which means they aren't experimentally achievable precisions yet > They give us the theoretical ceiling, but we still have noise and limitations in the lab.
Lev: And to wrap up on my side, this paper sets targets for next generation arrays based on a single M level state and gives us those ultimate quantum limits based on ideal resources > It’s a good roadmap for what experimentalists should aim for next.
Kai: So we're leaving with the idea that the multiparameter quantum bounds for entanglement-assisted aperture synthesis are governed by how many photons you collect and what kind of state you use > It's a lot to take in, but it really shows the potential power of these entangled arrays.
The paper's summary: Kai: So, we've got this paper on multiparameter quantum bounds for entanglement-assisted aperture synthesis, and what it boils down to is that they're looking at how much better we can image an extended scene when you use entangled light across multiple stations.
Mira: Right. They take the standard problem of just measuring one baseline visibility and turn it into this bigger, trickier thing where you have to estimate all the visibilities simultaneously, which is a multiparameter problem.
Lev: And they tackle that by using this reduction lemma, which basically cuts the complex optical setup down to looking at the quantum Fisher information of just one delocalized photon.
Kai: So, what they find is that for a single photon state with entanglement supplied, you get a fraction of that information—specifically r over one plus r where r is the number of pairs supplied to a baseline. That's the first big piece of math there.
Mira: But then they immediately hit this roadblock when they calculate the QFI matrix and its mean Uhlmann curvature, because numerically it shows that for an extended source, it’s incompatible with measurement—that is, you can't get everything at once.
Lev: So they are saying that no single measurement optimized for one parameter can actually achieve the best possible result across all parameters at the same time.
Kai: They quantify this incompatibility by showing that their Holevo bound exceeds the symmetric logarithmic derivative bound by about seventy-four percent, which is a pretty big gap.
Mira: That means even with ideal quantum resources, you're still losing a chunk of potential precision because of how measurements work.
Lev: Before we get to how they try to beat these limits, they compare the performance of an explicit collective receiver against a global mode-sorting measurement and a pairwise beam-splitter receiver.
Kai: And that comparison is pretty telling because the collective receiver gets within about seven percent of the symmetric logarithmic derivative bound for two points on a six station array.
Mira: That's significantly better than what those pairwise receivers are doing, which is roughly an order of magnitude worse.
Lev: This experiment really shows that using a collective measurement setup is way better than just processing the individual baselines separately.
Kai: Then they talk about resource allocation, where they figure out when the whole array protocol actually makes sense by calculating this noisy-resource advantage threshold.
Mira: They find a break-even baseline around twenty kilometers for near-term parameters, which gives us a scale for what's feasible right now.
Lev: And they also found an optimal allocation strategy using a closed form square root law solution, which favors concentrating the entanglement budget on the lowest visibility baselines.
Kai: So if you only have limited entanglement, this rule tells you exactly where to put it to get the best result in terms of advantage.
Mira: The conclusion is that quantum imaging precision scales linearly with the number of photons collected and it's governed by a single M level state.
Lev: But they also make sure to be clear that these are ultimate quantum bounds assuming ideal resources, so we aren't talking about what we can actually build in the lab yet.
Kai: So they give us these theoretical limits based on ideal resources, which sets the target for what next generation arrays should aim for.
The paper's improvements: Tom: So, we're looking at how the authors suggest ways to make this multiparameter estimation problem even more tractable, and what that means for future experiments on these quantum arrays.
Kai: The main idea is that they're aiming to move from just theoretical bounds to actually designing a concrete system setup based on how much entanglement you have.
Mira: They suggest treating the whole thing as a collective multiparameter estimation problem, which they claim makes it much more manageable because it shows that the imaging precision scales linearly with the number of photons you collect and it's governed by a single M level state.
Lev: That's good, but we need to check if that linear scaling actually holds up when you factor in the overhead of error correction, because real hardware isn't perfect.
Kai: Exactly. And they point out that for allocating the entanglement budget itself, they found a closed form square root law solution, which is optimal only within the independent baseline family.
Mira: That rule means if you have limited entanglement resources, you should concentrate it on the lowest visibility baselines first because that’s where it gives you the biggest bang for your buck.
Lev: So what this means for building hardware is that we don't need to waste resources trying to optimize every single baseline individually; we just focus our entanglement budget there.
Kai: They also established a noisy-resource advantage threshold, which tells you when the whole array protocol becomes worthwhile, setting a break-even baseline around twenty kilometers for near-term parameters.
Mira: That threshold is crucial because it defines the regime where the array protocol actually pays off compared to just using individual stations.
Lev: So if we're building this, we have a clear distance and resource limit before we start designing arrays that are too big or too complex for current error correction overhead.
Kai: The authors also provide these ultimate quantum bounds, but they make it very clear that these are not experimental precisions because they assume ideal resources.
Mira: That's fair, they aren't claiming we can hit those numbers right now with the hardware we have; those are just the theoretical limits if everything worked perfectly.
Lev: So what this means for us is setting targets for the next generation of quantum arrays based on these single M level states and these ultimate quantum limits, even if we can't reach them yet.
Conclusion: Tom: So to wrap things up, this paper on "Multiparameter quantum bounds for entanglement-assisted aperture synthesis" really shows that we can get much better imaging precision by using entangled light across multiple stations than we could with just looking at one baseline visibility.
Kai: It's a lot to take in because the core finding is that the precision scales linearly with how many photons you collect and it's governed by a single M level state, which gives us some concrete targets for next generation arrays.
Mira: I think what this means for the world is that it sets a theoretical ceiling based on ideal resources, even if we can't build systems that hit those numbers yet with current hardware.
Lev: From an engineering standpoint, the main thing is knowing exactly what to build toward when you allocate your entanglement budget because they gave us that square root law solution.
Kai: I agree with Lev, it’s a practical prescription for building these things, concentrating the entanglement on the low visibility baselines where it helps most.
Mira: And before we move on to other papers, just remember that these ultimate quantum bounds are based on ideal resources which means our current noise models and detector inefficiencies are still missing from this analysis.
Lev: It does mean we still have a lot of work ahead regarding loss and detector inefficiency because the authors flagged those limitations specifically for future treatments.
Kai: So, while we can use these results to guide our designs, the next step is definitely taking this noise model and adding in things like finite repeater rates and more realistic loss scenarios.
Mira: That sounds like a solid plan for future work, but it's important to remember that this paper is a first step in understanding the full picture of what entanglement can do for extended scene imaging.
Kalaga Madhav, *, Pieter Kok, *, Nic Scott
Astrophotonics (innoFSPEC) · Leibniz-Institut für Astrophysik Potsdam (AIP) · School of Mathematical and Physical Sciences, The University of Sheffield · The CHARA Array of Georgia State University
quant-ph, astro-ph.IM, physics.optics
Submitted: 2026-09-09
Updated: 2026-11-18
Comments: 12 pages, 5 figures, 2 tables. Accepted for publication in Quantum Science and Technology
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 83/100
The gist: Entanglement-assisted optical interferometry promises diffraction-limited imaging over baselines where light cannot be physically combined…
Key concepts
- Multiparameter Quantum Estimation Problem
- This is the challenge of jointly estimating many different parameters (like visibilities on all baselines) simultaneously. The paper shows that treating this as a single, collective problem makes it mathematically tractable for quantum estimation.
- Local Photon-Number Superselection Rule (SSR)
- This rule causes a local photon to lose all phase information when measured individually. However, when photons are entangled across the array, shared entanglement allows some of this lost phase information to be recovered.
- Noisy-Resource Advantage Threshold
- This threshold defines the baseline distance where using the full quantum array protocol becomes beneficial compared to classical methods. The break-even point is around 20 km for near-term parameters, indicating when the quantum advantage starts to pay off.
Terminology
Summary
Entanglement-assisted optical interferometry promises diffraction-limited imaging over baselines where light cannot be physically combined, but existing theory treats only a single visibility on one two-station baseline The work formulates the multiparameter quantum estimation problem for an M-station array imaging an extended scene This research provides concrete design targets by recasting quantum aperture synthesis as collective multiparameter estimation. p. 0 de
Keywords: quantum metrology, multiparameter estimation, quantum Fisher information, Holevo bound, optical
interferometry, aperture synthesis, entanglement-assisted telescopes
Abstract
Entanglement-assisted optical interferometry promises diffraction-limited imaging over
baselines where light cannot be physically combined, but existing theory treats only a
single visibility on one two-station baseline. This work formulates the multiparameter
quantum estimation problem for an M-station array imaging an extended scene.Where the paper says it
The Problem and Reduction Lemma
The paper addresses the challenge of estimating complex visibilities on all M baselines jointly, which is intrinsically a multiparameter problem Theorem 1 reduces this multimode optical problem to the QFI of a single delocalized photon The local photon-number superselection rule (SSR) erases all phase information from the bare state, while shared entanglement restores a fraction f(r) = r/(1 + r) of the QFI on a baseline supplied with r pairs.
Multiparameter Incompatibility
The computation of the QFI matrix and its mean Uhlmann curvature reveals numerically that a point source is measurement-compatible, whereas for every extended-source model examined it is incompatible The Holevo bound exceeding the symmetric-logarithmic-derivative (SLD) bound by up to ∼74%. This incompatibility R > 0 means that no single measurement optimized parameter-by-parameter attains the multiparameter optimum. p. 3 74
(∼32% in the error bars on fitted parameters). Operationally, no single measurement optimized
parameter-by-parameter attains the multiparameter optimum.Where the paper says it
Receiver Performance Comparison
The study explicitly compares an explicit collective receiver, a global mode-sorting (quantum Fourier transform) measurement, against an explicit pairwise beam-splitter receiver For a two-point source on a generic M = 6 array, the collective receiver attains a weighted parameter variance within ∼7% of the SLD bound, whereas the pairwise receiver is roughly an order of magnitude worse. This explicit demonstration shows that a collective receiver outperforms pairwise visibility processing. p. 4 from being jointly optimal. This is an explicit demonstration that a collective receiver outperforms
pairwise visibility processing, complementing the measurement-independent statement of Sec.Where the paper says it
Resource Thresholds and Allocation
The noisy-resource advantage threshold delimits the regime where the array-level protocol is worthwhile, with a break-even baseline of ∼20 km for near-term parameters The optimal allocation of a finite entanglement budget is found to be a convex program with a closed-form proportional (square-root-law) solution favoring low-visibility baselines. This rule concentrates entanglement on the lowest-visibility baselines. p. 5 Applied to CHARA’s baselines,
Eq. (12) concentrates entanglement on the lowest-visibility baselines (the most resolved).Where the paper says it
Conclusion and Future Directions
The paper concludes that quantum-enhanced imaging precision scales linearly with collected photons and is governed by a single M-level state The results set targets for the next generation of quantum-enhanced arrays. The analysis provides ultimate quantum bounds (QFI/Holevo limits assuming ideal resources), which are not experimentally achievable precisions. p. 5 7) schedules a finite entanglement budget, reinforcing the faintest fringes. Throughout, these are ultimate quantum bounds (QFI/Holevo limits assuming ideal resources),
not experimentally achievable precisions. The noise model of Sec. 6 is a first step; a full treatment
of loss, detector inefficiency and dark counts, mode mismatch, finite-fidelity entanglement, memory
decoherence, and finite repeater rate (App.Where the paper says it
Where the paper says it
--- Page 8 ---
References
[3] Gottesman D, Jennewein T and Croke S 2012 Phys. Rev. Lett. 109 070503 doi:10. p. 7 2597311
[2] Mandel L and Wolf E 1995 Optical Coherence and Quantum Optics (Cambridge: Cambridge
University Press)
[3] Gottesman D, Jennewein T and Croke S 2012 Phys. 109 070503
doi:10. 109 070503
doi:10. 1103/PhysRevLett. 109 070503
doi:10. 1103/PhysRevLett. 109 070503
doi:10. 1103/PhysRevLett. 109 070503
doi:10. 1103/PhysRevLett.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
[4] Khabiboulline E T, Borregaard J, De Greve K and Lukin M D 2019 Phys. Rev. Lett. 123 070504 doi:10. p. 7 070503
[4] Khabiboulline E T, Borregaard J, De Greve K and Lukin M D 2019 Phys. 123
070504 doi:10.Where the paper says it
[5] Khabiboulline E T, Borregaard J, De Greve K and Lukin M D 2019 Phys. Rev. A 100 022316 doi:10. p. 7 070504
[5] Khabiboulline E T, Borregaard J, De Greve K and Lukin M D 2019 Phys. A 100 022316
doi:10. A 100 022316
doi:10. 1103/PhysRevA.Where the paper says it
Where the paper says it
[6] Marchese M M and Kok P 2023 Phys. Rev. Lett. 130 160801 doi:10. p. 7 022316
[6] Marchese M M and Kok P 2023 Phys. 130 160801
doi:10.Where the paper says it
[7] Brown M R, Allgaier M, Thiel V, Monnier J D, Raymer M G and Smith B J 2023 Phys. Rev. Lett. 131 210801 doi:10. p. 7 160801
[7] Brown M R, Allgaier M, Thiel V, Monnier J D, Raymer M G and Smith B J 2023 Phys. 131 210801 doi:10.Where the paper says it
[8] Stas P-J et al 2026 Nature 651 326 doi:https://doi.org/s41586-026-10171-w
[9] Tsang M 2011 Phys. Rev. Lett. 107 270402 doi:https://doi. p. 7 1038/s41586-026-10171-w
[9] Tsang M 2011 Phys. 107 270402 doi:10. 109 070503
doi:10. 1103/PhysRevLett. 109 070503
doi:10. 1103/PhysRevLett. 109 070503
doi:10. 1103/PhysRevLett.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
[10] Zhang Y and Jennewein T 2025 Phys. Rev. Research 7 043278 doi:https://doi. p. 7 270402
[10] Zhang Y and Jennewein T 2025 Phys. Research 7 043278 doi:10.Where the paper says it
[11] Modak S and Kok P 2025 Phys. Rev. A 111 043701 doi:https://doi. p. 7 1103/bf51-tj3j
[11] Modak S and Kok P 2025 Phys. A 111 043701 doi:10. A 100 022316
doi:10. 1103/PhysRevA. A 100 022316
doi:10. 1103/PhysRevA. A 100 022316
doi:10. 1103/PhysRevA.Where the paper says it
Where the paper says it
Where the paper says it
Where the paper says it
[12] Huang Z, Baragiola B Q, Menicucci N C and Wilde M M 2024 Phys. Rev. A 109 052434 doi:https://doi. p. 7 043701
[12] Huang Z, Baragiola B Q, Menicucci N C and Wilde M M 2024 Phys. A 109 052434
doi:10.Where the paper says it
[13] Wang Y, Zhang Y and Lorenz V O 2025 Phys. Rev. Research 7 023154 doi:https://doi. p. 7 052434
[13] Wang Y, Zhang Y and Lorenz V O 2025 Phys. Research 7 023154
doi:10. Research 7 023154
doi:10. 1103/PhysRevResearch.Where the paper says it
Where the paper says it
[14] Braunstein S L and Caves C M 1994 Phys. Rev. Lett. 72 3439 doi:https://doi. p. 7 023154
[14] Braunstein S L and Caves C M 1994 Phys. 72 3439
doi:10.Where the paper says it
[15] Paris M G A 2009 Int. J. Quantum Inf. p. 7 3439
[15] Paris M G A 2009 Int. Quantum Inf. Quantum Inf. 7 125 doi:10.Where the paper says it
Where the paper says it
[16] Carollo A, Spagnolo B, Dubkov A A and Valenti D 2019 J. Stat. Mech.: Theory Exp. p. 7 1142/S0219749909004839
[16] Carollo A, Spagnolo B, Dubkov A A and Valenti D 2019 J. : Theory Exp. : Theory Exp. 2019
094010 doi:10.Where the paper says it
Where the paper says it
[17] Ragy S, Jarzyna M and Demkowicz-Dobrzański R 2016 Phys. Rev. A 94 052108 doi:https://doi. p. 7 1088/1742-5468/ab3ccb
[17] Ragy S, Jarzyna M and Demkowicz-Dobrzański R 2016 Phys. A 94 052108
doi:10.Where the paper says it
[18] Holevo A S 1982 Probabilistic and Statistical Aspects of Quantum Theory (Amsterdam: North-Holland)
[19] Albarelli F, Friel J F and Datta A 2019 Phys. Rev. Lett. 123 200503 doi:https://doi. p. 7 052108
[18] Holevo A S 1982 Probabilistic and Statistical Aspects of Quantum Theory (Amsterdam:
North-Holland)
[19] Albarelli F, Friel J F and Datta A 2019 Phys. 123 200503
doi:10.Where the paper says it
[20] ten Brummelaar T A, McAlister H A, Ridgway S T, Bagnuolo Jr. W G, Turner N H, Sturmann L, Sturmann J, Berger D H, Ogden C E, Cadman R, Hartkopf W I, Hopper C H and Shure M A 2005 Astrophys. p. 7 200503
[20] ten Brummelaar T A, McAlister H A, Ridgway S T, Bagnuolo Jr. W G, Turner N H,
Sturmann L, Sturmann J, Berger D H, Ogden C E, Cadman R, Hartkopf W I, Hopper C H
and Shure M A 2005 Astrophys. W G, Turner N H,
Sturmann L, Sturmann J, Berger D H, Ogden C E, Cadman R, Hartkopf W I, Hopper C H
and Shure M A 2005 Astrophys. 628 453 doi:10.Where the paper says it
Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI system can perform multiparameter estimation for extended scenes by treating it as a collective multiparameter estimation problem, using
Theorem 1 shows quantum-enhanced imaging precision scales linearly with collected photons and is governed by a single M-level state, making the multiparameter problem tractable.
-
Improved AI system can design optimal entanglement allocation strategies for quantum arrays by applying the
closed form square-root law
solution which isoptimal only within the independent-baseline family,
thereby concentrating resources on low-visibility baselines. -
Improved AI system can determine when an array protocol is worthwhile by calculating a
noisy-resource advantage threshold,
which delimits, through a break-even baseline,when the array-level protocol is worthwhile.
-
Improved AI system can perform explicit collective receiver measurements to achieve performance near the Holevo bound, as the paper shows that
the collective receiver attains a weighted parameter variance within ∼7% of the SLD bound.
Abstract
Entanglement-assisted optical interferometry promises diffraction-limited imaging over baselines where light cannot be physically combined, but its estimation-theoretic bounds exist only for a single visibility on one baseline. This work formulates the multiparameter quantum estimation problem for an M-station array. A reduction lemma shows that the multimode quantum Fisher information (QFI) equals the mean photon number times the QFI of a single delocalized photon, making the problem finite-dimensional. The local photon-number superselection rule erases all phase information from the bare state; shared entanglement restores a fraction f(r)=r/(1+r) of the QFI on a baseline supplied with r pairs, a benchmark that applies to every pair of the array. The QFI matrix and its mean Uhlmann curvature show numerically that a point source is measurement-compatible, whereas every extended-source model examined is incompatible; a semidefinite-programming evaluation shows that the Holevo bound exceeds the symmetric-logarithmic-derivative bound by up to 24% for three stations and by 4-8% for larger arrays and for CHARA. An explicit collective receiver, a numerically optimised global linear-optical network with photon counting, attains a weighted variance within 2-9% of the Holevo bound, over an order of magnitude below an explicit pairwise receiver, and a noisy-resource threshold gives the break-even baseline (22 km for illustrative parameters) beyond which distributed entanglement beats direct transmission. Finally, allocating a finite entanglement budget has a closed-form proportional (square-root-law) solution for a per-baseline fringe cost, whereas for the source-parameter variance under an explicit photon budget the routing of photons matters far more than the entanglement split. These results recast quantum aperture synthesis as collective multiparameter estimation, illustrated on CHARA.
Related papers
- Computational regimes in matrix-product-state-based quantum trajectory simulations
- Toward chemically accurate quantum simulations on quantum hardware: A quantum-HPC hybrid approach
- Transversal surface-code game with reconfigurable qubits
- Fulqrum: a generalized framework for quantum subspace diagonalization
- Flux-Activated Resonant Control of a Bosonic Quantum Memory
- Machine learning the effects of many quantum measurements