Security of quantum key distribution with iterative sifting
Security of quantum key distribution with iterative sifting
复制标题
DOI：
10.1088/2058-9565/aa89bd
复制
发表时间：
2016-10
影响因子：
6.7
通讯作者：
K. Tamaki;H. Lo;Akihiro Mizutani;G. Kato;C. Lim;Koji Azuma;M. Curty
中科院分区：
文献类型：
--
作者：
K. Tamaki;H. Lo;Akihiro Mizutani;G. Kato;C. Lim;Koji Azuma;M. Curty
几种量子密钥分发（QKD）协议采用迭代筛选。在每一轮量子传输之后，Alice和Bob公开他们针对检测到的信号的部分设置信息（包括他们的基础选择）。然后，当满足基依赖终止条件时，即，每个基的检测信号的数量超过某些预先商定的阈值。然而，最近Pfister等人（2016 New J. Phys. 18 053001）表明，基依赖终止条件使得QKD不安全，特别是在有限密钥体制中，他们建议在完成量子阶段后公开所有设置信息。然而，这个协议有两个主要的缺点：它需要爱丽丝拥有一个大的内存，她还需要有一些关于量子信道的传输速率的先验知识。在这里，我们解决了这两个问题，通过引入一个基地独立的终止条件，在有限的密钥制度的迭代筛选。结合Azuma不等式使用该条件提供了对需要应用的隐私放大量的精确估计，从而导致QKD协议的安全性，包括具有迭代筛选的容损协议（Tamaki等人2014 Phys.Rev.A90052314）。我们的分析表明，在每一轮量子传输后公布基础信息不会损害容错协议的密钥生成速率。我们的研究结果允许实现更广泛的类的经典后处理技术在量子密钥分配与量化的安全性。
Several quantum key distribution (QKD) protocols employ iterative sifting. After each quantum transmission round, Alice and Bob disclose part of their setting information (including their basis choices) for the detected signals. This quantum phase then ends when the basis dependent termination conditions are met, i.e., the numbers of detected signals per basis exceed certain pre-agreed threshold values. Recently, however, Pfister et al (2016 New J. Phys. 18 053001) showed that the basis dependent termination condition makes QKD insecure, especially in the finite key regime, and they suggested to disclose all the setting information after finishing the quantum phase. However, this protocol has two main drawbacks: it requires that Alice possesses a large memory, and she also needs to have some a priori knowledge about the transmission rate of the quantum channel. Here we solve these two problems by introducing a basis-independent termination condition to the iterative sifting in the finite key regime. The use of this condition, in combination with Azuma’s inequality, provides a precise estimation on the amount of privacy amplification that needs to be applied, thus leading to the security of QKD protocols, including the loss-tolerant protocol (Tamaki et al 2014 Phys. Rev. A 90 052314), with iterative sifting. Our analysis indicates that to announce the basis information after each quantum transmission round does not compromise the key generation rate of the loss-tolerant protocol. Our result allows the implementation of wider classes of classical post-processing techniques in QKD with quantified security.