Researchers at the University of Hong Kong have extended fully passive security, eliminating source-side hacking vulnerabilities, to multi-party quantum communication via a protocol called Quantum Conference Key Agreement (CKA). Building on a recent development in measurement-device-independent QKD, the new interference-based CKA layers security measures to address vulnerabilities on both the source and detector sides of quantum communication. This approach achieves good key rate performance, particularly in high-loss channels, through the use of single photon interference. The work demonstrates a high level of implementation security for many-user communication.

Interference-Based CKA & Measurement-Device-Independence

This advancement addresses a critical gap in secure communication by providing a method for multiple users to establish a shared key, a feat previously complicated by the potential for source-based attacks. The team’s approach layers security measures, building upon measurement-device-independent QKD, which already defends against detector-side hacking, with the newly implemented fully passive techniques. The newly developed protocol uses single photon interference to maintain strong key rate performance even in communication channels experiencing significant signal loss.

This resilience is achieved through a phase mis-matching strategy, a technique employed to improve sifting efficiency, a common challenge in fully passive protocols, particularly when scaling to multiple users. The team found that while enjoying the benefit of better security at the sources, not much key rate performance needs to be sacrificed. Numerical simulations, facilitated by a high-performance integration library and an optimized computer program, were essential in validating the protocol’s efficacy.

Addressing the computational demands of simulating a four-party fully passive CKA required innovative techniques; the researchers introduced a “branch cutting” method to increase computation efficiency without compromising key rate performance. This optimization allowed for detailed analysis of practical imperfections, including fluctuations in phase measurements, intensity variations and non-uniform phase distribution. The foundation of this work lies in combining a recently developed interference-based prepare-and-measure quantum CKA with the advantages of fully passive QKD.

Fully passive QKD schemes eliminate side channels originating from the source modulation, a vulnerability present in many traditional QKD systems. The authors note the growing interest in this approach to security. The team’s innovation extends this concept to a multi-user CKA, enabling secure communication among a larger group of participants.

Unlike active CKA protocols that require parties to alternate between key generation and parameter estimation rounds, the fully passive CKA performs all experiments in a uniform manner, simplifying the process and potentially reducing complexity. This simplification is achieved by eliminating the need for basis selection, a common feature in traditional QKD protocols. The researchers detail that adjustments to the yield estimation from decoy state analysis are necessary due to the unique characteristics of the fully passive channel model.

Extending Fully Passive QKD to Multi-User Conferences

Researchers in Hong Kong have extended the idea of fully passive Quantum Key Distribution (QKD) to a multi-party communication scenario, a step beyond many existing protocols focused on two-party exchanges. This work combines the idea of fully passive QKD with CKA, which could offer advantages from both ideas. By implementing a phase mis-matching strategy, the researchers were able to improve sifting efficiency, allowing for reasonable performance in a multi-user setting.

The researchers also introduced a “branch cutting” method, a computational technique that enhances computational speed. Analysis of the generated intensities revealed minimal self-correlation, suggesting a similar lack of correlation among phases, which is essential for secure key generation. The researchers state that this improvement is an important step toward practical implementation of fully passive CKA.

Phase Mis-Matching Strategy Improves Sifting Efficiency

A newly implemented phase mis-matching strategy significantly improves the efficiency of sifting key material in multi-user quantum conference key agreement (CKA), addressing a longstanding challenge for fully passive quantum communication systems. This advancement allows for the use of a broader range of signals, circumventing the heavy sifting that typically plagues such protocols and hinders key rate performance. Computational demands posed a significant hurdle to simulating this fully passive CKA system, given the extensive high-dimensional integration calculations required.

To overcome this, the team developed a high-performance integration library coupled with an optimized computer program. Further enhancing computational speed, they introduced a method to streamline calculations without compromising the resulting key rate. This efficient computation is critical for assessing the protocol’s performance under realistic conditions and validating its potential for practical implementation. The researchers demonstrate that this approach yields a good sifting efficiency and reasonable performance.

Computational Optimization for Fully Passive CKA Simulation

This optimization was coupled with the introduction of a “branch cutting” method, a technique designed to increase computational speed without diminishing the resulting key rate performance. The “branch cutting” method partially alleviates the computational burden associated with calculating key rates in fully passive CKA systems, effectively streamlining the process.

The team accounted for practical imperfections in the simulation, including a two percent misalignment for each receiver and a phase-measurement fluctuation with a standard deviation of five degrees, to assess performance under more realistic conditions. Further improvements in key rate performance are anticipated with access to more powerful computing resources, specifically through optimization of the slice number and input intensity, and the complete calculation of all slice choices without employing the “branch cutting” method.

The team plans to investigate the scaling relationship between the number of users and the sifting effect introduced by fully passive sources, indicating a pathway toward expanding the system’s capacity. This future work will build on the current findings, aiming to determine how the computational demands and key rate performance scale with an increasing number of participants in the quantum conference.

Modeling Practical Imperfections in the Protocol

Simulating realistic conditions for this new protocol required accounting for fluctuations in phase measurement, intensity and the uniformity of phase distribution across the system. The team detailed these considerations in Sections IV and V of their published work, focusing on how these imperfections impact key generation rates. Explicitly modeling these factors allowed for a simulated assessment of a four-party fully passive CKA system, a departure from many prior studies concentrating on two-party communication.

A key advantage of this protocol lies in the elimination of basis sifting between users, as each participant simply adopts the publicly announced basis after signal transmission. This simplification is particularly beneficial as the number of users increases, streamlining the key generation process. The protocol uses a shared physical signal for both key generation (KG) and privacy enhancement (PE) rounds, differentiating them through post-processing steps completed after transmission.

This design choice circumvents the need for complex basis reconciliation procedures common in traditional quantum key distribution. The team for the first time applied the fully passive scheme to a multi-user scenario, along with the application of a phase mis-matching strategy, results in a good sifting efficiency and reasonable performance. The team’s calculations focused on corrected yields and precise phase error rate estimations to refine the accuracy of the simulated key rate.

The security analysis for this fully passive CKA mirrors that of active CKA protocols, with the primary distinction being the imperfections inherent in the fully passive source preparation. Instead of assuming perfectly prepared signals, the simulations allowed for phase variations within a range of [-Δ,Δ], where Δ is defined as π/M.

This realistic modeling approach acknowledges that real-world quantum devices are not ideal and that these imperfections must be accounted for to ensure secure communication. The team’s work provides an important step toward realizing practical, secure multi-party quantum communication networks capable of operating even with imperfect hardware.

Four-Party CKA Results & Key Rate Calculation

Researchers detailed these findings while investigating practical imperfections such as phase and intensity fluctuations, and non-uniform phase distribution, within the system. The team’s work, outlined in Section IV of their paper, focused on assessing performance under realistic conditions, moving beyond idealized theoretical models. While this initially results in poor performance, the researchers used a phase mis-matching strategy to use all signals and improve the overall key rate.

Further analysis using a two-decoy setting revealed an additional key rate reduction when calculating yield upper bounds. The team acknowledges this as an area for future development. An experimental realization of this fully passive, many-user system is planned as a subsequent step, and the team anticipates this will be the first demonstration of its kind.

👉 More information
🗞 Fully passive quantum conference key agreement
✍️ Jinjie Li, Wenyuan Wang and H. F. Chau
🧠 DOI: http://link.aps.org/doi/10.1103/6bq6-s1tw

More like this