⚛️ quantum physics

Entanglement-swapping measurements for deterministic entanglement distribution

This paper characterizes a family of projective entanglement-swapping measurements built from complex Hadamard operators that enable deterministic, order-independent entanglement distribution with optimal G-concurrence for all pure input links, thereby eliminating the need for outcome-based postselection.

Original authors: Mir Alimuddin, Jaemin Kim, Antonio Acín, Leonardo Zambrano

Published 2026-08-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Mir Alimuddin, Jaemin Kim, Antonio Acín, Leonardo Zambrano

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

In the emerging field of quantum networking, scientists are trying to build a new kind of internet that uses the strange rules of quantum physics to send information. The most valuable resource on this network is entanglement, a deep connection between two particles that persists even when they are far apart. Imagine two coins that, no matter how far apart they are, always land on the same side when flipped; this is the essence of the connection, though in reality, it involves much more complex properties than simple coin flips. To send this connection over long distances, researchers cannot simply send a single particle across a continent, because the signal fades and gets corrupted by the environment. Instead, they must build a chain of stations. Each station holds a piece of the connection and passes it along to the next, effectively stitching together short links into one long, unbroken thread. This process of stitching is called entanglement swapping.

The challenge with this stitching process has always been its unpredictability. When a station performs the measurement required to connect two links, the result is random. In many cases, different random results lead to different final states of connection. Some results might create a very strong link, while others create a weak one, or a link with a different quality entirely. To fix this, traditional methods often require the network to discard the weak or different results and try again, hoping for a "good" outcome. This wastes time and resources, as the energy and particles used in the failed attempts are thrown away. The goal for engineers has been to find a way to make the process deterministic, meaning that no matter which random result occurs, the final connection is always the same, or at least can be easily corrected to be the same, without ever needing to throw anything away.

A team of researchers has now solved this problem for a wide range of quantum networks. They have identified a specific set of measurement techniques that guarantee every possible outcome of the swapping process produces an identical final connection, up to a simple adjustment that the end users can make. This means the network never has to reject a result. Furthermore, they proved that these specific techniques do not just produce identical results; they produce the best possible results. For any two links being connected, no other measurement method can create a stronger final connection on average, and crucially, this maximum strength is achieved in every single instance, not just as a statistical average.

The researchers worked out the mathematical rules that define these perfect measurements. They found that the measurements must be constructed from a special type of pattern, known in mathematics as a complex Hadamard matrix. These patterns are like grids of numbers with very specific symmetries. The team discovered that for the simplest networks, involving two-level systems, there is essentially only one way to build these perfect patterns. For slightly more complex three-level systems, there is also just one way. However, as the systems grow larger, the number of ways to build these patterns changes dramatically. For five-level systems, there are exactly seventy-two distinct families of these patterns. For systems with four levels, and any size that is a multiple of four, there are not just many, but an infinite number of ways to construct them. This classification is vital because it tells engineers exactly which tools they can use to build a reliable network.

The study also looked at how these perfect measurements hold up when the real world gets in the way. In a perfect lab, the particles are pure, but in reality, they are often mixed with noise. The researchers showed that if the noise is a specific type of random disturbance, the measurements still work perfectly, producing identical results that can be corrected. Even when the noise is more general and unpredictable, the average quality of the connection remains very close to the ideal, with the error staying within a tight, predictable limit. This robustness suggests that these methods are not just theoretical ideas but are practical enough for real-world deployment.

Another critical finding concerns the order in which the network operates. In a long chain of stations, one might wonder if it matters which station connects its links first. For the simplest networks, the researchers proved that the order does not matter at all; the final result is the same regardless of the sequence of operations. This gives network designers great flexibility, as they do not need to coordinate a strict schedule for every station. However, they also found that this freedom has a limit. In systems with four levels or higher, the order of operations can change the final result, meaning that for more complex future networks, the sequence of connections will need to be carefully planned.

By providing a complete map of these measurement techniques, the researchers have removed a major bottleneck in the design of quantum networks. Their work eliminates the need to discard failed attempts, ensuring that every bit of energy and every particle used in the network contributes to the final goal. They have shown that it is possible to distribute the strongest possible quantum connections with absolute certainty, turning a process that was once a gamble of chance into a reliable, deterministic engineering task. This advancement brings the vision of a global quantum internet, capable of secure communication and powerful computing, significantly closer to reality.

Technical Summary: Entanglement-Swapping Measurements for Deterministic Entanglement Distribution

Problem Statement
Entanglement swapping is a fundamental primitive for distributing entanglement across quantum networks. However, for generic projective measurements and pure input states, different measurement outcomes typically yield end-to-end states with distinct Schmidt spectra (entanglement structures). This necessitates branch-dependent processing or the rejection of unfavorable outcomes (postselection) to achieve a target state. Since entanglement monotones cannot increase on average under local operations and classical communication (LOCC), discarding outcomes trades success probability for output quality, wasting consumed links. The paper addresses the challenge of identifying measurement schemes where every outcome produces an entangled state that is locally unitary (LU) equivalent to a common target, thereby eliminating the need for postselection while maintaining optimal entanglement distribution.

Methodology
The authors analyze a standard three-party swapping scenario involving Alice (A), a swapping node (N), and Bob (B). Alice and Bob share arbitrary pure bipartite states $|\tilde{\psi}\rangle$ and $|\tilde{\phi}\rangle$ of local dimension $d$ with the node.

Alignment: The protocol assumes nodes apply local unitary rotations to align the incoming states into a common Schmidt basis (computational basis), resulting in diagonal coefficient matrices $A$ and $B$ .
Measurement Representation: The node performs a rank-one projective measurement $M_N = \{|\Gamma_i\rangle\langle\Gamma_i|\}_{i=1}^{d^2}$ . The measurement vectors are represented via coefficient matrices $E_i$ such that $|\Gamma_i\rangle = (E_i^* \otimes \mathbb{1})|\Phi\rangle$ , where $|\Phi\rangle$ is the unnormalized maximally entangled state.
Constraints: The study restricts attention to measurements where all vectors have full Schmidt rank (nonsingular $E_i$ ).
Optimization Metric: The authors utilize G-concurrence ( $C_d$ ), defined as $d |\det M|^{2/d}$ for a state with coefficient matrix $M$ , as the scalar entanglement monotone to evaluate optimality.

Key Contributions and Results

1. Characterization of Universal LU-Determinism
The paper defines Universal LU-determinism as the property where, for every pair of pure input states, all measurement outcomes yield output states that are LU-equivalent (differing only by local unitaries correctable based on the outcome).

Necessary Conditions: For a protocol to be universally LU-deterministic, the measurement outcome probabilities must be uniform ( $p_i = 1/d^2$ ), and the coefficient matrices $E_i$ must belong to the set of unbiased measurement operators (SUBM), where all entries have modulus $1/d$ .
Necessary and Sufficient Conditions: The authors prove that two operators $E_i, E_j \in \text{SUBM}$ generate LU-equivalent outputs for all inputs if and only if they belong to the same Phase-Conjugation (PC) class. Specifically, $E_i$ and $E_j$ must be related by:
$E_i = D_L E_j D_R \quad \text{or} \quad E_i = D_L E_j^* D_R$
where $D_L$ and $D_R$ are diagonal unitary matrices.

2. Optimality and Complex Hadamard Measurements
The authors optimize the average G-concurrence. They establish that a measurement is optimal (maximizing average G-concurrence for all inputs) if and only if all measurement vectors are maximally entangled.

Combining optimality with universal LU-determinism restricts the coefficient matrices to scaled complex Hadamard unitaries ( $E_i = \frac{1}{\sqrt{d}}H_i$ ).
Crucially, for these measurements, every individual outcome attains the optimal G-concurrence, not just the ensemble average. This eliminates the trade-off between success probability and output quality.
The paper demonstrates that standard Bell measurements are not universally LU-deterministic in the Schmidt-aligned basis but can be converted into such measurements via a fixed local Fourier pre-rotation, followed by outcome-dependent phase corrections.

3. Classification of PC-Classes
The authors classify the distinct PC-classes of complex Hadamard operators based on dimension $d$ :

$d = 2, 3$ : There is exactly one PC-class.
$d = 5$ : There are exactly 72 distinct PC-classes.
$d = 4k$ (where $k$ is a positive integer): There are uncountably many PC-classes.
$d = 4$ : Specifically noted as having uncountably many classes.
$d \geq 6$ (not divisible by 4): The classification remains an open problem.
The authors note that while different PC-classes yield outputs with the same optimal G-concurrence, the resulting states are generally incomparable under deterministic LOCC.

4. Robustness and Network Composition

Noise Robustness: Exact LU-determinism is preserved under independent bipartite depolarizing noise if the measurement operators are related by diagonal unitaries without complex conjugation. For general convex input contamination ( $\rho = (1-\epsilon)\rho_{\text{ideal}} + \epsilon\sigma_{\text{noise}}$ ), the outcome-averaged trace distance from the ideal output is bounded by the contamination weight $\epsilon$ .
Order Independence: In swapping chains, the authors show that for $d=2$ and $d=3$ , the corrected end-to-end state is independent of the order in which swapping operations are performed (associativity). For $d=4$ , they provide a counterexample showing that order independence is not guaranteed, although G-concurrence optimality is maintained for each swap.

Significance and Claims
The paper claims to provide a complete characterization of all projective swapping measurements with full-Schmidt-rank vectors that are universally LU-deterministic.

Postselection-Free Distribution: The proposed schemes allow for deterministic entanglement distribution where every measurement outcome is accepted, eliminating the waste of links associated with rejecting unfavorable branches.
Individual Optimality: Unlike previous works that optimized ensemble averages, these schemes guarantee optimal G-concurrence for every single outcome.
Scalability: The framework supports embedding these modules into arbitrary network topologies (chains, lattices) with precise guarantees on robustness and composition.
Dimensional Constraints: The work highlights a sharp distinction in the complexity of deterministic swapping between low dimensions ( $d=2,3$ ) and higher dimensions, particularly regarding the uniqueness of the measurement class and the associativity of swapping operations.

The authors conclude that their results establish a rigorous foundation for postselection-free, G-concurrence-optimal entanglement swapping in pure-state quantum networks, with specific guarantees on robustness and the algebraic properties of swapping chains in low dimensions.

Technical Summary: Entanglement-Swapping Measurements for Deterministic Entanglement Distribution

More like this