Until now, proving security in prepare-and-measure quantum key distribution needed detailed modelling of devices; a team at Université Libre de Bruxelles (ULB) and Lund University has bypassed this requirement. Bounding information leakage from Alice’s device, the probability of guessing her input, guarantees secure communication against collective attacks. Their best protocols achieve one secret bit per round under ideal conditions and tolerate visibilities down to 0.84, equivalent to an 8% error rate.

A new approach simplifies proving secure communication in quantum key distribution by focusing on limiting information leaks from Alice’s device rather than modelling its detailed operation. Bounding how easily an eavesdropper could guess Alice’s input is enough to guarantee secrecy against attacks attempting to intercept data.

This approach centres around a “leakage bound,” imagining you’re whispering a secret; this sets a limit on how clearly someone eavesdropping can hear it. They achieved these results by using approximations in complex calculations, akin to rounding off numbers for easier mental math, to find solutions that are “good enough” even if not perfect.

Reduced visibility tolerances enable leakage resilient secure communication

Visibilities as low as 0.84, coupled with a quantum bit error rate of 8%, are now tolerated by new protocols. This represents an improvement over standard device-dependent BB84 implementations that previously required an 11% threshold for secure communication. Proving security against collective attacks no longer requires detailed source modelling, but instead relies on measuring how much information about Alice’s input escapes through emitted signals or side channels.

The long-standing requirements of prepare-and-measure QKD, where accurate device modelling has proven difficult and introduced vulnerabilities, are bypassed in this approach. Further analysis reveals classical information content limits: Gcl values range from 2√2 for both (4,3) and (4,4) protocols up to 3 log₂3 for (6,3).

These calculations also determined white noise visibility thresholds vcl and detection efficiency ηcl, the lowest being 1/√2 for both (4,3) and (4,4), indicating strong resistance against signal degradation and loss. While these numbers demonstrate progress in tolerating imperfections, realistic conditions with complex detector vulnerabilities or atmospheric disturbances still pose a substantial hurdle to widespread deployment.

Quantifying laboratory leakage enables near-term secure quantum networks

Real-world deployment immediately faces the engineering challenge of accurately quantifying information leakage from Alice’s laboratory. Sophisticated measurement techniques beyond simple characterisation of emitted states are needed to establish precise limits on escaping data through intended signals or unintended “side channels”. Although certifying that this leakage remains within acceptable bounds may compromise key generation rates or require new protocols, it sidesteps the need for detailed device modelling. This work’s importance isn’t diminished by limitations in precisely quantifying leakage; instead, a practical path forward is offered for quantum communication systems.

The team demonstrated security assuming only a limit on information escaping Alice’s lab, a less demanding requirement than fully modelling her device. This could enable simpler hardware implementations while strengthening security against eavesdropping and providing a foundation for secure networks over the coming decade. With visibilities down to 0.84, one secret bit per round can be achieved, an improvement over previous thresholds where established B92 and BB84 protocols would fail under such minimal conditions. By focusing on bounding information leakage, specifically how easily an eavesdropper could deduce Alice’s input from her laboratory, researchers have simplified security proofs for quantum communication. Limiting this alone guarantees protection against collective attacks without needing detailed modelling of internal device components; exhaustive characterisation, a requirement of existing protocols like BB84, introduces vulnerabilities.

This research demonstrated that security in quantum key distribution can be proven by limiting the amount of information leaking from the sender’s lab, rather than requiring complete knowledge of the devices used. This approach offers a simpler path to secure communication because it reduces reliance on complex and potentially vulnerable hardware modeling. The team showed that with visibilities down to 0.84, corresponding to an 8% error rate, secret keys could still be generated using modified protocols. Researchers are now focused on accurately quantifying this leakage in real-world systems as a necessary step for implementation.

👉 More information
🗞 Prepare-and-measure quantum key distribution from a bound on information leakage
✍️ Abhishek Mishra, Michele Masini, Maria Balanzó-Juandó and Stefano Pironio (Université libre de Bruxelles); Armin Tavakoli (Lund University)
🧠 ArXiv: https://arxiv.org/abs/2610.37296

More like this