自检测型量子密钥分配研究

Traditional quantum key distribution (QKD) protocols, such as BB84, must assume that the quantum devices of communication peers are fully characterized by themselves before the security may be guaranteed. However, the security proofs of self-testing QKD (ST-QKD) protocols have less assumptions on their devices implementations, and thus ST-QKD has a high level of security and is becoming a very hot topic recently. There are many important issues to be studied in the area of ST-QKD, such as how to achieve fully ST-QKD (FST-QKD) protocol is still a challenge to present technology, and it is proved that the security of measurement-device ST-QKD (MST-QKD) may need to be strengthened. To implement practical QKD systems with higher security level, we will carry out our research in the following three respects. 1. In this project, the research group plans to study how to achieve FST-QKD by photon amplification technique. FST-QKD is the QKD protocol with highest security level so far, but not practical to present technology. Photon amplification technique may be the most feasible way to implement FST-QKD experimentally. 2. To implement MST-QKD with higher security level or better feasibility will be also studied in this project. The security of original MST-QKD protocol relies on the accuracy of photon transmitter's encoding operation. We plan to carry out new MST-QKD protocol experiment whose security is not related to the accuracy of encoding and thus is more secure. We also focus on MST-QKD implementations combined with modified coherent sources and reference-frame-independent approach, which lead to MST-QKD with longer distance and higher key bit rate. 3. In theory study, our aim is to devise new ST-QKD protocol, which has fewer assumptions on security than BB84, and also more feasible than FST-QKD protocol. The technique is to focus on the combination of MST-QKD and two-way QKD. The project covers the security and feasibility issues, which are essential problems in the research of ST-QKD. For FST-QKD protocol, which is the QKD protocol with highest security level so far, our aim is to make it more practical and can be achieved in lab at least. For MST-QKD, we focus not only on how to strengthen its security but also how to implement it with better performance. Therefore, this project has important academic significance to the study of quantum cryptography.

自检测型量子密码系统的安全性对其内部实现依赖较少，因而具有很高的安全性，深受学术界关注。但其一些重要问题仍待研究，如怎样实现完全自检测型协议仍旧是一个难题，而测量设备自检测协议的安全性仍有增强的潜力。本项目计划对自检测型量子密码开展实验为主兼有理论的全面研究。课题组计划以光子放大方案为主要技术途径来研究完全自检测型协议的实现技术；以实现安全性对编码误差免疫的测量自检测协议为目标，开展更安全的测量设备自检测协议的实现研究；以修改相干态光源和参考系无关方案为两条路径研究实用性更好的测量设备自检测协议。在理论方面，课题组则研究构造安全性优于传统BB84协议，可用性又优于完全自检测协议的新型量子密钥分发协议。本项目涵盖了自检测型协议的主要研究热点问题即安全性问题和可用性问题：对于安全性最高的完全自检测协议，以其实现方法为研究目标；对测量设备自检测协议，则在安全性和可用性两方面开展研究。

不同于安全性非常依赖于实现细节的传统量子密钥分发系统，自检测型量子密钥分发系统的安全性对其内部实现细节依赖较少，因而具有很高的安全性，近年来深受学术界关注。然而，在实际条件下，以测量设备无关协议为代表的自检测型量子密钥分发协议的应用还受到很多限制。本项目以自检测型协议的安全性和实用性为研究目标，在量子密钥分发的相关研究领域开展了深入的理论和实验研究，四年来共发表SCI收录学术论文43篇，其中一区论文4篇，取得了一系列具有一定学术影响力的工作。下面简述部分重要研究成果。1.首次实现了无需相位参考系校准的测量设备无关系统，简化了实际系统运行的流程，显著提升了该类系统的环境鲁棒性，结果发表在物理学顶级期刊物理评论快报上。后续进一步改进这一系统，将主动扰偏引入该系统使其不仅保持了免相位参考系校准运行的优点，还使其免信道偏振扰动的影响，实现了真正意义上的环境扰动鲁棒型系统，成果发表在光学顶级期刊Optica上。2.首次实现了安全性不依赖于编码态制备精度的测量设备无关实验系统，解决了长期以来编码误差对该类系统安全性的影响，显著改善了实际系统的安全性。这项成果发表在光学权威学术期刊光学快报上。3.基于课题组提出的基于全概率表计算随机数小熵的方法，实验实现了一个自检测型量子随机数产生器，原则上可在大损耗条件下产生量子随机数，这项工作发表在光学快报上。该成果可用于今后的随机数产生器设计，对于自检测型QKD也有启发意义。4.在新型量子密钥分发协议的研究方面，首次将相位随机化与窃听者辅助态退相干结合，解决了准确估计高维编码协议信息泄露量的难题，完善并发展了高维编码协议设计和分析方法，并进行了实验验证，论文发表在顶级期刊自然通讯上。本项目注重人才培养和学术交流，培养国家优秀自然科学青年基金获得者1人，博士毕业生3人，其中1人获国家奖学金，资助课题组多位成员参加国内外学术交流活动。