arXiv News

First-hand information for everyone

Site

AboutPrivacy PolicyTerms of Use

© 2026 arXiv News

arXiv News

EnglishJapanese

Switch language

EnglishJapanese
Loading account…

First-hand information for everyone

Back to front page
Quantum PhysicsEnglishPublished

Researchers Show Ideal Entanglement‑Swapping Protocol Cannot Be Trusted, Propose Probabilistic Check Instead

November 22, 2026arXiv: 2611.26436v1

Scientists have proved that a perfect entanglement‑swapping service, where a middle party (Charlie) would hand over a faithfully entangled pair to two remote users, can never be composably secure. Even if Charlie’s quantum measurement is accepted as correct, a dishonest Charlie can simply rename his measurement outcome and learn nothing about the state that Alice and Bob actually receive. The study argues that adding ordinary noise to the envisioned network does not fix this flaw unless the noise level itself is strictly capped.

To address the shortfall, the researchers designed a cut‑and‑choose protocol. Charlie prepares many pairs, and the protocol randomly selects a subset of these pairs for joint measurement in two complementary bases, Z and X. Alice and Bob test whether the outcomes show the expected anti‑correlation. If the fractions of anti‑correlated results stay below pre‑agreed thresholds, the protocol accepts the remainder of the pairs; otherwise it aborts.

When the protocol accepts, the accepted set of pairs is guaranteed to contain at most a tiny fraction of misreported pairs. The probability that the misreport rate exceeds the desired target is bounded by a user‑chosen security parameter, ε_σ. Mathematically, the authors prove that acceptance directly limits the misreport rate and guarantees a lower bound on the fidelity of the delivered entangled state.

However, the authors emphasize a key limitation: a deliberate misreport by Charlie looks exactly like honest read‑out noise in his own Bell measurement. Therefore no protocol can distinguish between a misreport and natural measurement noise during operation; they can only certify that the total error stays below a minimum noise level agreed upon beforehand. This establishes a quality‑of‑service guarantee rather than true individual‑pair security.

The study demonstrates the fundamental impossibility of an ideal noiseless entanglement‑swapping service and offers a practical, probabilistic certification method that balances security with operational feasibility.

Source paper

When lies look like noise: certifying untrusted entanglement swapping

Quantum Physicscs.DC
View on arXiv
Researchers Show Ideal Entanglement‑Swapping Protocol Cannot Be Trusted, Propose Probabilistic Check Instead | arXiv News