🔬 optics

Multiparameter quantum bounds for entanglement-assisted aperture synthesis

This paper establishes a multiparameter quantum estimation framework for entanglement-assisted optical interferometry, demonstrating that shared entanglement restores phase information lost to superselection rules, enables collective measurements to significantly outperform pairwise strategies for extended sources, and provides a closed-form solution for optimally allocating entanglement resources to maximize imaging performance on arrays like CHARA.

Original authors: Kalaga Madhav, Pieter Kok, Nic Scott

Published 2026-09-10

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Kalaga Madhav, Pieter Kok, Nic Scott

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

To see the surface of a distant star or the swirling gas around a black hole, astronomers need telescopes with mirrors as wide as a city. Since building a single mirror of that size is impossible, they use a technique called interferometry, linking many smaller telescopes together to act as one giant instrument. The light from each telescope must be brought to a central point to be combined, but over long distances, the air and the fiber optic cables that carry the light destroy the delicate phase information needed to form an image. For decades, this physical limitation has kept the resolution of optical telescopes from reaching its theoretical maximum. A new approach suggests using quantum entanglement—a connection between particles that persists even when they are far apart—to bypass the need for physical cables. Instead of sending the starlight itself, the telescopes share entangled particles that act as a bridge, allowing them to measure the star's properties without ever bringing the light together.

Until now, the theory behind this quantum bridge was only worked out for two telescopes looking at a single point of light. Real astronomy, however, involves many telescopes arranged in an array, trying to image complex, extended objects like the surface of a star or a binary system. A team of researchers has now expanded this theory to cover the full complexity of a multi-telescope array. They found that while the math for a single pair of telescopes is simple, the problem of imaging a whole scene with many telescopes is fundamentally different. Their work shows that to get the best possible image, the telescopes cannot simply measure each pair of connections independently. Instead, they must process the information from all telescopes together as a single, collective event.

The researchers demonstrated that when dealing with extended sources, the information carried by the light is distributed in a way that makes it impossible to extract the full picture by looking at each telescope pair one by one. If astronomers tried to measure the connections between telescopes individually, they would lose a significant amount of detail, with the potential error in their measurements increasing by up to 74 percent compared to the theoretical limit. To overcome this, the team proposed a specific measurement strategy where the signals from all telescopes are sorted and analyzed simultaneously using a global quantum operation. This collective approach, which treats the entire array as a single machine, was shown to recover nearly all the lost information, performing far better than any method that treats the telescope pairs separately.

The study also addressed the practical reality that quantum resources are noisy and imperfect. Entanglement is fragile, and sending it over long distances introduces errors. The researchers calculated a specific threshold to determine when using this quantum bridge is actually better than simply trying to send the light through a fiber optic cable. They found that for near-term technology, the quantum method becomes advantageous once the distance between telescopes exceeds about 20 kilometers. Beyond this point, the quantum link provides more useful information per photon than a direct physical connection, even when accounting for the losses and imperfections in the system. This gives engineers a clear target for where to deploy these new technologies.

Finally, the team figured out how to best distribute a limited supply of entangled particles across the many connections in a telescope array. Intuition might suggest giving more resources to the strongest connections, but the math shows the opposite is true. To get the sharpest image, the entangled particles should be concentrated on the telescope pairs that have the weakest signals. This strategy, which prioritizes the most difficult measurements, ensures that the entire array works at its highest possible efficiency. By framing the problem as a collective estimation task rather than a series of individual measurements, this work provides a concrete blueprint for the next generation of quantum-enhanced telescopes, turning a theoretical possibility into a design plan for seeing the universe with unprecedented clarity.

Technical Summary: Multiparameter Quantum Bounds for Entanglement-Assisted Aperture Synthesis

Problem Statement
Optical/near-infrared very-long-baseline interferometry (OLBI) is limited by the inability to physically combine light from distant apertures without destroying phase information. Quantum-assisted interferometry offers a solution by distributing entanglement to act as a nonlocal optical link, allowing visibility extraction without co-locating photons. However, existing theoretical frameworks are restricted to single-baseline, single-visibility estimation (scalar problems). Real-world imaging of extended sources requires the joint estimation of complex visibilities across all $\binom{M}{2}$ baselines of an $M$ -station array, constituting a multiparameter estimation problem. This work addresses the gap between single-baseline theory and the full array estimation problem, specifically investigating the quantum limits, measurement incompatibility, and resource allocation for entanglement-assisted aperture synthesis.

Methodology
The authors formulate the problem using quantum metrology theory, focusing on the Quantum Fisher Information (QFI) and the Holevo Cramér–Rao bound.

State Reduction: The optical field from an incoherent source is modeled as a weak thermal state with mean photon number $\epsilon M \ll 1$ . The state is truncated to the single-excitation sector. A Reduction Lemma (Theorem 1) is proven, demonstrating that the multimode QFI matrix for the array state $\rho$ scales linearly with the total photon number and is equivalent to the QFI of a single delocalized photon in a normalized state $\sigma = G/M$ , where $G$ is the coherence matrix. This collapses the infinite-dimensional multimode problem into a finite-dimensional one.
Superselection Rules (SSR) and Entanglement: The analysis accounts for the local photon-number superselection rule, which erases phase information from local detection. The authors quantify how shared entanglement restores accessible information. For a baseline $(i,j)$ supplied with $r$ pairs of single-photon-entangled ancillas, the accessible phase QFI is restored by a factor $f(r) = r/(1+r)$ .
Multiparameter Analysis: The authors compute the QFI matrix ( $J$ ) and the mean Uhlmann curvature ( $U$ ) to determine the scalar incompatibility parameter $R = \|2iJ^{-1}U\|_\infty$ . This parameter dictates the gap between the Symmetric Logarithmic Derivative (SLD) bound and the Holevo bound, indicating whether optimal measurements for all parameters can be performed jointly.
Receiver Comparison: Two explicit receiver architectures are compared:
- Collective Receiver: A global mode-sorting measurement (discrete Quantum Fourier Transform) across all $M$ stations.
- Pairwise Receiver: Standard pairwise beam-splitter measurements for each baseline, analogous to classical aperture synthesis.
Resource Optimization: The paper models a noisy-resource advantage threshold comparing entanglement-assisted links against direct transmission through lossy channels. It further derives the optimal allocation of a finite budget of entangled pairs ( $R_E$ ) across baselines to minimize reconstruction variance.

Key Contributions and Results

Reduction to Single-Photon QFI: The work establishes that the multiparameter QFI for an $M$ -station array imaging an extended scene is governed entirely by the QFI of a single delocalized photon state $\sigma$ . This provides a tractable framework for analyzing large arrays.
Measurement Incompatibility: Numerical analysis reveals a fundamental distinction between source types:
- For a point source, the state is pure, $R=0$ , and the SLD bound is attainable; per-baseline processing is optimal.
- For extended sources (e.g., binary stars), the state is mixed, and $R > 0$ . For the models examined, $R$ peaks near 0.74, meaning the Holevo bound can exceed the SLD bound by up to ~74%. This implies that no single measurement optimized parameter-by-parameter can attain the multiparameter optimum; joint processing across baselines is necessary.
Receiver Performance: The explicit collective receiver (global QFT) achieves a weighted parameter variance within ~~7% of the SLD bound. In contrast, the explicit pairwise receiver is approximately an order of magnitude worse (~~88% higher variance), demonstrating a concrete advantage for collective measurements beyond pairwise visibility processing.
Noisy-Resource Threshold: A break-even baseline length ( $L^\star$ ) is derived where entanglement-assisted interferometry outperforms direct transmission. For near-term parameters (repeater efficiency $\eta_e=0.5$ , Bell fidelity $F=0.95$ ), $L^\star \approx 20$ km. Beyond this distance, repeater-distributed entanglement yields higher Fisher information per stellar photon than direct fiber transmission.
Optimal Entanglement Allocation: The optimal distribution of a finite entanglement budget follows a "proportional" (square-root) law: $r^\star_{ij} \propto \sqrt{w_{ij}}/\gamma_{ij}$ . This strategy favors allocating more entanglement to baselines with lower visibility (higher resolution requirements), rather than a "water-filling" approach used for capacity maximization.

Significance and Scope
The paper recasts quantum aperture synthesis as a collective multiparameter estimation problem, providing concrete design targets for next-generation quantum-enhanced arrays, illustrated using the CHARA array.

The authors explicitly frame these results as ultimate quantum bounds (assuming ideal resources) rather than experimentally achievable precisions. The work serves as a theoretical foundation, identifying the fundamental limits imposed by measurement incompatibility and the specific advantages of collective receivers.

The paper maintains a modest scope regarding future work, noting that:

The collective receiver presented is near-optimal (within ~7% of the SLD bound) but does not provably attain the Holevo bound.
The noisy-resource analysis currently models only link loss and Bell-pair fidelity, excluding detector inefficiency, mode mismatch, and memory decoherence.
The incompatibility results are numerical for specific source models rather than universal theorems.
The allocation rule assumes independent per-baseline resources; correlated multipartite entanglement is left for future investigation.

The primary contribution is the formulation of the multiparameter problem, the proof of the reduction lemma, and the demonstration that extended-source imaging inherently requires joint measurements to approach fundamental quantum limits.

More like this