Certification of quantum correlations and DIQKD at arbitrary distances through routed Bell tests
Pavel Sekatski, Jef Pauwels, Edwin Peter Lobo, Stefano Pironio, Nicolas Brunner · 2025
Read Certification of quantum correlations and DIQKD at arbitrary distances through routed Bell tests by Pavel Sekatski, Jef Pauwels, Edwin Peter Lobo, Stefano Pironio, Nicolas Brunner (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-2502-12241.txt.
Introduction
Pavel Sekatski, Jef Pauwels, Edwin Peter Lobo, Stefano Pironio, and Nicolas Brunner show in two thousand twenty-five how to certify quantum correlations over any distance despite signal loss. Device-independent tests prove quantumness without trusting the hardware, but photons lost in transit have capped their range. That cap also limits device-independent key distribution.
Their trick uses routed Bell tests, where a particle can be measured near the source or far away. They prove that tests with only entangled qubit pairs can still certify quantum correlations under arbitrary loss on the long channel. The proof adapts ideas from self-testing and quantum steering. They then sketch a key distribution protocol that stays secure, in principle, over arbitrary distances. The routing idea sidesteps the loss barrier.
Catalog record
| CREATOR | Pavel Sekatski, Jef Pauwels, Edwin Peter Lobo, Stefano Pironio, Nicolas Brunner |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 103 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-12241
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Certification of quantum correlations and DIQKD at arbitrary distances through routed Bell tests Pavel Sekatski, Jef Pauwels, Edwin Peter Lobo, Stefano Pironio, Nicolas Brunner Abstract Transmission loss represents a major obstacle to the device-independent certification of quantum correlations over long distances, limiting applications such as device-independent quantum key distribution (DIQKD). In this work, we investigate the recently proposed concept of routed Bell experiments, in which a particle sent to one side can be measured either near or far from the source. We prove that routed Bell tests involving only entangled qubits can certify quantum correlations even in the presence of arbitrary loss on the channel to the distant device. This is achieved by adapting concepts from self-testing and quantum steering to the routed Bell test framework. Finally, as a natural extension of our approach, we outline a DIQKD protocol that, in principle, is secure over arbitrary distances. Device-independent (DI) protocols aim to certify quantum properties—such as entanglement or security—without making assumptions about the internal functioning of the measurement devices [ 1 ] . At the core of any DI claim lies a fundamental classical-quantum gap that can be established solely from the observed correlations. In addition, such claims rely on a specific causal structure—typically that of a Bell experiment—in which spatially separated parties perform local measurements on a shared quantum state. Within this framework, the classical-quantum gap manifests as a violation of a Bell inequality and is synonymous with the notion of quantum nonlocality [ 2 ] . In practice, harnessing the power of quantum nonlocality comes with significant experimental challenges. In particular, local detection efficiencies must exceed a specific threshold; otherwise, a local model might exploit the so-called “detection loophole” to explain the observed correlations [ 3 , 4 ] . The detection efficiency required to violate a Bell inequality varies considerably with the experimental setup. While Bell tests involving many measurement settings and high-dimensional entangled states can, in principle, tolerate very low efficiencies [ 5 , 6 , 7 , 8 , 9 , 10 , 11 ] , practical implementations typically rely on low-dimensional systems such as qubits. In these scenarios, the efficiency thresholds remain relatively high—typically above 60%—even when loss is the only imperfection [ 2 ] . For applications such as device-independent quantum key distribution (DIQKD), the requirements are even more stringent, with standard protocols becoming insecure below 50% efficiency [ 12 , 13 ] . Although recent experiments have achieved loophole-free Bell violations [ 14 , 15 ] and proof-of-principle demonstrations of DIQKD [ 16 , 17 ] , distributing nonlocal correlations over long distances remains a major challenge. Transmission losses—which effectively reduce the overall detection efficiency—increase exponentially with distance, imposing severe limitations even when local detectors operate at unit efficiency. This has motivated the exploration of heralded entanglement generation [ 18 , 19 , 20 , 21 ] , though such approaches remain out of reach with current technology. Figure 1: The routed Bell scenario . A source distributes entanglement to Alice and Bob. A switch positioned after the source directs Bob’s system either to a nearby measurement device {\rm B}^{\rm S} , or to a distant device {\rm B}^{\rm L} , which experiences loss with transmission \eta . We show that the observation of a Bell violation between {\rm A} and {\rm B}^{\rm S} allows the certification of genuine quantum correlations along the long path, for arbitrary transmission \eta>0 . Recently, an alternative approach to mitigating transmission loss in Bell experiments– coined “routed Bell tests”–was proposed [ 22 , 23 ] . In this scenario, illustrated in Fig. 1 , a standard bipartite Bell experiment is modified so that one of the particles (e.g., Bob’s) can be measured at two different locations: either near the source or at a distant site. Interestingly, these tests can tolerate more loss than standard Bell tests [ 23 ] , and can increase robustness to loss for DIQKD [ 24 , 25 ] . The setup is closely related to DI protocols with a “local Bell test” [ 26 ] . In the present paper, we prove that routed Bell tests using entangled qubits can tolerate arbitrarily high transmission loss, enabling the distribution of genuine quantum correlations over an arbitrarily large distance. Our approach is analytical, drawing concepts from self-testing [ 27 , 28 ] and quantum steering [ 29 , 30 , 31 ] . In particular, we derive a tight family of steering inequalities for qubits, which may be of independent interest. As an extension of our analysis, we outline a DIQKD protocol for the routed Bell scenario, which can in principle achieve a non-zero key rate at arbitrary distances. The routed Bell scenario.— In a routed Bell test [ 22 , 23 ] , illustrated in Fig. 1 , a source prepares a bipartite quantum state \rho_{AB} , sending subsystem A to Alice –who is positioned close to the source – and subsystem B toward Bob. Upon receiving an input x , Alice performs a measurement described by a set of positive operator-valued measures (POVMs) \{\text{A}_{a|x}\}_{x} and obtains an outcome a . Bob’s particle is routed by a switch to one of two measurement stations, depending on an input r\in\{S,L\} . If r={\rm S} (for “short path” or SP), the system is measured at a nearby device {\rm B}^{\rm S} ; if r={\rm L} (for “long path” or LP), it is measured at a distant device {\rm B}^{\rm L} . The measurements at SP and LP are described by POVMs \{{\rm B}^{\rm S}_{b|y}\}_{y} and \{{\rm B}^{\rm L}_{b|y}\}_{y} , respectively, where y is the input and b the output. These measurements may differ between the two locations. The correlations observed in the experiment are given by p(a,b|x,y,r)=\tr(\rho_{AB}\A_{a|x} \otimes\B^r_{b|y})\,. (1) In the DI spirit, no assumptions are made about the internal workings of the source, the measurements, or the switch—all are treated as untrusted. The only physical assumptions are causality constraints enforcing the tensor product structure in Eq. ( 1 ), and the independence of the devices. For example, Alice’s measurement process is assumed to be independent of the inputs and outputs on Bob’s side (which includes the switch and both measurement stations), and vice versa. In the routed Bell scenario, the fact that the measurement devices {\rm A} and {\rm B}^{\rm S} are located close to the source reduces losses and noise, making it easier to observe a Bell violation in the SP. Such a violation allows for the partial certification of key properties of the source and Alice’s measurement device. The central goal is then to certify genuine quantum correlations in the LP between {\rm A} and {\rm B}^{\rm L} , possibly leveraging the information provided by the SP violation. Unlike standard Bell tests and traditional device-independent (DI) protocols, our aim is not to exclude a classical explanation for the entire experiment—this is already achieved via the Bell violation in the SP. Instead, we seek to rule out a broader class of hybrid quantum-classical models, in which the devices near the source may operate quantumly, while only classical information reaches {\rm B}^{\rm L} . Correlations of this kind are referred to as short-range quantum (SRQ), following Ref. [ 23 ] . SRQ correlations