Background: Quantum networks can not only break through the physical limits of traditional classical networks to achieve unconditionally secure information transmission but also demonstrate revolutionary application values in fields such as quantum computing, quantum sensing, and precision metrology. As one of the core applications of quantum networks, the quantum conference key agreement (QCKA) protocol allows multiple users in a network to share a common secret key. By enabling secure multipartite encryption, QCKA offers robust application prospects for confidential data transmission and secure financial infrastructure.

However, current QCKA protocols face significant challenges in practical implementation. On one hand, traditional QCKA protocols rely on the distribution of multipartite entangled states (such as GHZ states). Nevertheless, preparing high-fidelity, high-brightness multipartite entangled states in experiments is extremely difficult, and the preparation efficiency of entangled states drops significantly as the number of users increases, which greatly limits the transmission distance and key rate of the protocol. Although measurement-device-independent (MDI) QCKA protocols offer stronger security and alleviate the requirements on light sources, the stringent demands for time synchronization and photon indistinguishability make it difficult to operate efficiently in practical networks. On the other hand, the scalability of previous experimental schemes is limited. As the number of users increases, the complexity of sources and related devices rises rapidly, making it hard to apply in large-scale networks. Therefore, the realization of efficient and scalable QCKA in practical quantum networks remains a major challenge.

Research Progress: To address this challenge, a collaborative team led by Professor Zeng-Bing Chen from Nanjing University and Associate Professor Hua-Lei Yin from Renmin University of China has experimentally demonstrated an efficient multiuser source-independent (SI) QCKA protocol based on Bell state distribution. As shown in Fig. 1, the team utilizes an untrusted central node to generate and distribute polarization-entangled photon pairs. All users independently and randomly select a measurement basis (Z basis or X basis) to perform projection measurements on the received photons. If Alice and a certain Bob select the same measurement basis and detect photons simultaneously, it is recorded as a valid event. By performing post-matching on the measurement results of valid events and applying corresponding XOR operations, equivalent multipartite quantum correlations (GHZ correlations) can be established among multiple users via classical correlations, thereby circumventing the technical difficulties of preparing and distributing multipartite entangled states. Finally, post-processing including error correction and privacy amplification is performed on the results to obtain the final secure conference key.

Compared with previous QCKA experiments, the scheme demonstrates advantages in several aspects. First, by leveraging post-matching and post-processing techniques to transform multiple sets of bipartite entangled states into multipartite entanglement, this scheme maintains quantum-enhanced security while using relatively simple entangled sources instead of complex multipartite entangled sources, significantly reducing experimental complexity. Second, when extending users, the scheme only requires the central node to distribute one more set of entangled photon pairs and add corresponding detection modules, without the need to change the overall architecture, which lowers the threshold for network expansion. Finally, as an entanglement-based protocol, the scheme is naturally immune to all side-channel attacks targeting the source and security loopholes caused by source imperfections, ensuring communication security under non-ideal source conditions. The team built a system containing two polarization light sources and four sets of projection measurement devices and experimentally realized this protocol in a three-user star quantum network. The experimental setup is shown in Fig. 2.

The team conducted a total of six sets of experiments to systematically investigate the effects of different channel transmission and basis-selection probabilities on the key rate. The experimental results are in high agreement with the theoretical simulations. With a Z-basis selection probability of 0.9, the secure key rates achieve 21.1, 4.72, and 1.72 kbps under channel transmission of 0.164, 0.0824, and 0.052, respectively.

Future Prospects: This study experimentally demonstrates the feasibility of building an efficient and scalable SI-QCKA system. In the future, as related technology matures, QCKA is expected to be integrated into fully connected fiber quantum networks by combining technologies such as dense wavelength division multiplexing. Consequently, this approach is expected to satisfy the secure communication requirements of large-scale user groups at a reduced resource cost. Moreover, it will accelerate the practical deployment of applications such as quantum-encrypted teleconferencing and multipartite confidential data transmission, driving the evolution of quantum secure communication from point-to-point links toward multi-user networked architectures.

The complete study is accessible via DOI:10.34133/research.1034