A double selection entanglement distillation-based state estimator
Joshua Carlo A. Casapao 1 , Ananda G. Maity 1 2 , Naphan Benchasattabuse 3 , Michal Hajdušek 3 , Akihito Soeda 4 5 , Rodney Van Meter 6 , and David Elkouss 1 · 2025
Read A double selection entanglement distillation-based state estimator by Joshua Carlo A. Casapao 1 , Ananda G. Maity 1 2 , Naphan Benchasattabuse 3 , Michal Hajdušek 3 , Akihito Soeda 4 5 , Rodney Van Meter 6 , and David Elkouss 1 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-24280.txt.
Introduction
Joshua Carlo Casapao and six co-authors fold a measuring tool into a quantum network task, in a 2025 paper. Quantum networks will need to check the quality of their shared entangled states, but separate checking protocols waste the very resource being checked. Their idea is to read the state from the by-products of a purification routine already running.
The routine, called double selection distillation, sharpens weak entangled pairs. Its success and failure counts, the authors show, quietly encode the parameters of the incoming pairs and of the sharpened ones. Estimates drawn from those counts need fewer resources than an earlier distillation-based estimator. The approach promises to let future quantum networks audit themselves while they work, rather than stopping to measure.
Catalog record
| CREATOR | Joshua Carlo A. Casapao 1 , Ananda G. Maity 1 2 , Naphan Benchasattabuse 3 , Michal Hajdušek 3 , Akihito Soeda 4 5 , Rodney Van Meter 6 , and David Elkouss 1 |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 40 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-24280
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A double selection entanglement distillation-based state estimator Thanks: This work was supported by the JST Moonshot R&D program under Grants JPMJMS226C. Joshua Carlo A. Casapao 1 , Ananda G. Maity 1 2 , Naphan Benchasattabuse 3 , Michal Hajdušek 3 , Akihito Soeda 4 5 , Rodney Van Meter 6 , and David Elkouss 1 Abstract With the advent of practical quantum communication networks drawing closer, there is a growing need for reliable estimation protocols that can efficiently characterize quantum resources with minimum resource overhead requirement. A novel approach to this problem is to integrate an estimator into an existing network task, thereby removing the need for an additional characterization protocol. In this work, we show that the measurement statistics of a double selection distillation protocol alone can be used to efficiently estimate the Bell-diagonal parameters of the undistilled states, as well as the resulting distilled states after additional post-processing. We also demonstrate that this novel estimator outperforms the previously proposed distillation-based estimator in terms of resource complexity. I Introduction Recent experimental progress is enabling quantum communication tasks over near-term quantum networks, paving the way toward the future quantum Internet [ 1 , 2 , 3 ] . Inevitably, it becomes critical to simultaneously develop robust and efficient methods that certify and characterize the resources used in these networks to ensure proper functioning [ 4 ] . There exist different methods with varying assumptions [ 4 ] ; some examples are partial/full tomographic reconstruction [ 5 , 6 , 7 ] , direct fidelity estimation [ 8 ] , and randomized benchmarking techniques [ 9 , 10 , 11 ] . It is important that these methods compensate for experimental realities and prioritize both resource efficiency and measurement complexity without sacrificing the amount of learnable information. Typically, characterization and certification procedures are conducted independently before implementing other network tasks to ensure the desired condition of the resources used. In contrast, we can consider another avenue in which a characterization method is integrated into an existing network task unrelated to characterization. We expect one such scheme to exploit the behavior of the underlying structure of a quantum resource under a network task, and then post-process any learnable information to obtain a partial or even full characterization. For example, in the realm of quantum error correction, estimators in [ 12 , 13 ] can use syndrome measurements of stabilizer codes to learn Pauli noise parameters without disturbing the encoded information. By integrating characterization within network tasks, potential methods similar to these can indirectly address the problem of resource consumption by effectively reducing the number of separate protocols to be implemented. Recently, it was shown in [ 14 , 15 ] that a disti llation-based state esti mator can be designed (dubbed Disti-Mator) to efficiently characterize the Bell-diagonal elements of the states prepared for probabilistic distillation protocols. In this particular case, the action of these distillation protocols on the Bell-diagonal elements is exploited. It was shown that partial state estimation can be a natural by-product of post-processing the measurement statistics from distillation. Thus, we can avoid an additional benchmarking procedure whenever distillation is a necessary network task. However, a limitation of the original construction in [ 15 ] is its reliance on three distinct distillation protocols to characterize the Bell-diagonal elements. This may be undesirable if a user wants a consistent supply of high-fidelity states for further network tasks or just wants to avoid reconfiguring a distillation setup during implementation, which directly affects timing coordination. In this work, we build on top of this previously developed distillation-based estimator. We show that a distillation-based estimator can be designed more compactly by only implementing a double selection distillation protocol and using its measurement statistics, thus introducing a double selection-based Disti-Mator. We prove that the measurement statistics obtained from this protocol are enough to efficiently characterize the Bell-diagonal elements of the undistilled states, as well as the distilled states after additional post-processing. Our results are relevant in the context of establishing end-to-end entanglement over real testbeds, given that the proposed estimator avoids a major temporal bottleneck of changing measurement bases for the purpose of estimation. We structure this work as follows. We first revisit the double selection protocol introduced in [ 16 ] that we use in our investigation. We then describe the strategy for estimating the Bell-diagonal coefficients, including the relevant concentration bounds. We further show that the resulting distilled states can also be characterized after successfully estimating the noisy input states. We also investigate the number of samples required to estimate these coefficients. We then show that the proposed estimator can outperform the previously proposed Disti-Mator in [ 15 ] relative to the sample complexity. Finally, we provide some numerical tests to assess the utility and feasibility of our estimator during implementation. II Double selection protocol We consider the double selection distillation protocol [ 16 ] for two parties, Alice and Bob, with half of the protocol illustrated in Figure 1 . The protocol is as follows: 1. Alice and Bob prepare and share three identical noisy Bell pairs, which we label \rho^{(0)} , \rho^{(1)} , \rho^{(2)} . 2. Both parties implement a bilateral CNOT on \rho^{(0)} and \rho^{(1)} , with the former as the control qubit. 3. Both parties implement a bilateral CNOT on \rho^{(1)} and \rho^{(2)} , with the latter as the control qubit. 4. They bilaterally measure \rho^{(1)} in the Z basis and \rho^{(2)} in the X basis. Then they classically communicate their measurement results with each other. 5. Postselection stage: if both confirm coincidences in the Z measurements and in the X measurements, then \rho^{(0)} is successfully distilled and is prepared for the next round. Otherwise, \rho^{(0)} is discarded. Before each subsequent round, the roles of the X and Z bases on the reference frames of the surviving states are inverted. It has been shown in [ 16 ] that by using two ancillary noisy Bell pairs for postselection, this protocol can remove both phase and bit flip errors more efficiently than the single selection protocols in [ 17 , 18 ] that only use a single ancillary pair. Additionally, double selection provides higher-fidelity states much faster after several rounds and can achieve yields comparable to single selection. {\lx@inpgf@ignorespaces{\rho^{(0)}}} {\lx@inpgf@ignorespaces{\rho^{(1)}}} {\lx@inpgf@ignorespaces{\rho^{(2)}}} A Z Z A X X Fig. 1: Double selection distillation protocol [ 16 ] between two users, where one half of the protocol local to one user is shown. The protocol is implemented with three noisy input states and bilateral quantum operations. The protocol itself is a probabilistic protocol (that is, a successful distillation of a noisy input state is not deterministic) that requires two-way classical communication between the users for postselecting Z and X measurement results. III Main results III-A Estimation strategy We propose that both the undistilled and distilled states in a double selection protocol can be estimated using the coincidence probabilities over a single distillation round . Here, we consider that the input states \rho=\overline{\rho}+\rho_{\text{off-diagonal}} , where \overline{\rho} is in Bell-diagonal form: \overline{\rho}(\mathbf{q})\equiv q_{1}\outerproduct{\Phi^+}{\Phi^+}+q_{2}\outerproduct{\Phi^-}{\Phi^-}\\ +q_{3}\outerproduct{\Psi^+}{\Psi^+}+q_{4}\outerproduct{\Psi^-}{\Psi^-}. (1) We describe \mathbf{q}\equiv(q_{1},q_{2},q_{3},q_{4}) as the set of convex coefficients that satisfy the normalization q_{1}+q_{2}+q_{3}+q_{4}=1 . We further assume that q_{1}>1/2 , reflecting a situation in which the desired distilled