Asymptotic Limits of Entanglement Distribution

Piotr Masajada, Aby Philip, Alexander Streltsov · 2026

Read Asymptotic Limits of Entanglement Distribution by Piotr Masajada, Aby Philip, Alexander Streltsov (2026) 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-2605-23443.txt.

Introduction

Masajada, Philip, and Streltsov pin down when quantum entanglement can survive long repeater chains with local processing. Entanglement survives at long range exactly when the channel admits a correctable subspace; without one, the transmitted state collapses exponentially fast toward separable states, and standard filtering cannot rescue it. Running parallel channels per link can fight the decay, but the count per link must grow at least logarithmically with station count. For depolarizing noise below threshold, that logarithmic bound is reachable.

Catalog record

CREATORPiotr Masajada, Aby Philip, Alexander Streltsov
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE61425
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.23443

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Asymptotic Limits of Entanglement Distribution Piotr Masajada, Aby Philip, Alexander Streltsov Abstract Reliable distribution of quantum entanglement over long distances is a central challenge in quantum information science, fundamentally limited by decoherence in noisy communication channels. In this work, we investigate the asymptotic limits of entanglement distribution across quantum networks utilizing intermediate repeater stations and local operations and classical communication (LOCC). We establish a strict dichotomy: the asymptotic preservation of entanglement over arbitrarily long distances is possible if and only if the underlying quantum channel admits a correctable subspace. For channels lacking such a subspace, we prove that the transmitted state converges exponentially fast to the set of separable states, rendering standard LOCC filtering insufficient. To counteract this exponential degradation, we analyze networks employing parallel channel uses per link. We derive a fundamental lower bound on the physical resource requirements, proving that for channels without a correctable subspace, the number of parallel channels per link must scale at least logarithmically with the number of intermediate stations to sustain a non-zero amount of entanglement. This theoretical limit serves as a stringent benchmark for quantum repeater architectures and underscores the necessity of advanced quantum error-correcting codes, such as qLDPC codes, which show promise in saturating this optimal resource scaling. Introduction. Quantum entanglement 20 is a fundamental resource for a wide array of quantum information processing tasks, including quantum key distribution 4 , quantum teleportation 2 , and distributed quantum computing 22 ; 27 . A central challenge in realizing these technologies is the reliable distribution of entanglement between distant parties. In realistic scenarios, quantum states must be transmitted through noisy quantum channels, which inevitably interact with the environment and induce decoherence. This continuous degradation strictly limits the amount of entanglement that can be shared and maintained between remote parties. In the standard entanglement distribution scenario 32 ; 37 ; 24 ; 41 two distant parties, Alice and Bob, have access to a noisy quantum channel \Lambda . Alice first prepares a bipartite quantum state \rho in her laboratory, keeping one subsystem and using the channel to transmit the other subsystem to Bob. Because the transmission is imperfect, the channel acts nontrivially on the particle sent to Bob, while the particle retained by Alice remains unchanged. As a result, the initially prepared state \rho is transformed into the final bipartite state \Lambda\otimes\openone[\rho] . Any entanglement initially present in \rho is generally degraded by the action of the channel, and the properties of \Lambda determine how much entanglement can ultimately be shared between the two parties. If the noisy channel \Lambda is used sequentially, the resulting state shared by Alice and Bob is given by (\Lambda\circ\cdots\circ\Lambda)\otimes\openone[\rho] . This setting is relevant, for example, in long-distance quantum communication through a chain of noisy transmission segments, where the quantum system effectively undergoes several consecutive applications of the same channel before reaching Bob. In general, the detrimental effects of noise accumulate with the number of channel uses, leading to a progressive degradation of the shared entanglement 25 . In many physically relevant situations, this entanglement eventually disappears altogether in the limit of infinitely many channel applications 35 . To counteract these effects, parties can employ entanglement distillation 5 ; 3 ; 10 ; 11 and filtering protocols 36 . Using local operations and classical communication (LOCC) 7 , they can extract high-quality entangled states from a larger ensemble of degraded ones. Additionally, the use of intermediate repeater stations can significantly enhance the feasibility of establishing entanglement over long distances 1 . In this work, we investigate the fundamental and asymptotic limits of long-distance entanglement distribution in the presence of repeater stations. The central question we address is whether there exists any LOCC-based strategy capable of preserving a nonzero amount of entanglement in the asymptotic limit, that is, over arbitrarily long distances. In this scenario, after each application of the noisy channel \Lambda , the parties—or the stations assisting the transmission—may apply an LOCC protocol F_{i} to the shared state before the next channel use. These operations can depend on previous measurement outcomes and therefore need not be identical at every step; see also Fig. 1 . Such intermediate processing stations can be viewed as local checkpoints that attempt to counteract, delay, or partially compensate for the progressive accumulation of noise. In a multipartite transmission architecture involving many channels operating in parallel, these intermediate nodes play the role of quantum repeater stations 1 ; 28 . If the channel \Lambda is used n times during the transmission process, the resulting transformation of an initial state \rho can be written as \Lambda_{n}[\rho]=(\Lambda\otimes\openone)\circ F_{n}\circ\cdots\circ(\Lambda\otimes\openone)\circ F_{1}\bigl[\rho\bigr]. (1) The central question we address is whether, for a given quantum channel \Lambda , there exists any strategy involving such intermediate stations and filtering operations F_{i} that can preserve any entanglement in the asymptotic limit or as n\rightarrow\infty . While various entanglement measures have been introduced in the literature, in what follows we focus on the entanglement of formation 18 . For pure bipartite states, this quantity coincides with the entanglement entropy, namely E_{\mathrm{f}}(\psi^{AB})=S(\psi^{A}), (2) where S(\rho)=-\mathrm{Tr}(\rho\log_{2}\rho) denotes the von Neumann entropy. For a mixed bipartite state \rho^{AB} , the entanglement of formation is defined as the minimal average entanglement entropy over all pure-state decompositions of \rho^{AB} , that is 40 , E_{\mathrm{f}}(\rho^{AB})=\min_{\{p_{i},\psi_{i}^{AB}\}}\sum_{i}p_{i}E_{\mathrm{f}}(\psi_{i}^{AB}), (3) where the minimum is taken over all ensembles \rho^{AB}=\sum_{i}p_{i}\ket{\psi_{i}}\!\bra{\psi_{i}}^{AB} . Our central question is therefore whether, for an appropriate choice of LOCC filters F_{i} , it is possible to retain a nonzero amount of entanglement in the asymptotic limit n\rightarrow\infty , that is, \lim_{n\rightarrow\infty}E_{\mathrm{f}}(\Lambda_{n}[\rho^{AB}])>0. (4) Figure 1: Entanglement distribution through a noisy channel with intermediate repeater stations. An initial bipartite state \rho^{AB} is shared between parties A and B . Subsystem A is transmitted through a sequence of n noisy channel uses \Lambda , which may be understood as propagation through n successive transmission segments with intermediate stations located between Alice and Bob. After each channel use, the parties, possibly assisted by these intermediate stations, may apply an LOCC filter F_{i} . The figure illustrates the resulting state in Eq. ( 1 ) for n=2 . Single-qubit channels. To begin, we investigate entanglement distribution through a single-qubit quantum channel \Lambda . We assume Alice holds a single-qubit memory that she aims to entangle with Bob’s system across a sequence of n transmission segments, each linked via the quantum channel \Lambda . We recall that the Choi state of \Lambda is defined as 8 \Gamma=(\Lambda\otimes\openone)(\phi^{+}), (5) where \ket{\phi^{+}}=(\ket{00}+\ket{11})/\sqrt{2} is a Bell state. We consider the scenario in which the intermediate filtering operations F_{i} are allowed to be stochastic local operations and classical communication (SLOCC) 12 rather than deterministic LOCC. This broader framework is important because probabilistic filters can, upon success, increase the entanglement of the post-selected state which is not possible with deterministic LOCC. The key question is thus whether this additional freedom provided by SLOCC can preserve a nonzero amount of entanglement after many sequential applications of the channel by conditioning on successful filtering events. In the following, we show that although SLOCC filters may temporarily improve the quality of the surviving states, this does not overcome the fundamental limitations of the channel: any such apparent entanglement preservation comes at

Read the complete text