1.School of Computer Science and Engineering, Central South University, Changsha 410083, China 2.Center for Quantum Science and Technology, Jiangxi Normal University, Nanchang 330022, China
Abstract:Compared with discrete variable quantum key distribution (DVQKD), continuous variable (CV) QKD has high security bit rate and other advantages, which, however, are slightly insufficient in secure transmission distance. In addition, the application of quantum catalysis has significantly improved the performance of Gaussian modulated (GM) CVQKD, especially in secure transmission distance. Recently, the application of quantum catalysis has significantly improved the performance of GM-CVQKD. However, whether it can be used to improve the performance of discrete modulated (DM) CVQKD protocol is still ambiguous. Therefore, a scheme of DM CVQKD protocol based on quantum catalysis is proposed in this paper to further improve the performance of the proposed protocol in terms of secure key rate, secure transmission distance and maximum tolerable noise. Our results show that under the same parameters, when the transmittance T introduced by quantum catalysis is optimized, the proposed scheme can effectively further improve the performance of QKD system compared with the original four-state modulation CVQKD scheme. In particular, when the tolerable excess noise is 0.002, the use of quantum catalysis can break the safe communication distance of 300 km with a key rate of 10–8 bits/pulse. However, if this noise is too large, the improvement in the effect of quantum catalysis on protocol performance will be restrained. In addition, in order to highlight the advantages of the use of quantum catalysis, the ultimate limit PLOB (Pirandola-Laurenza-Ottaviani-Banchi) bound of point-to-point quantum communication is given in this paper. The simulation results indicate that although neither the original scheme nor the proposed scheme can break the bound, compared with the former, the latter can be close to the boundary in long-distance transmission. These results provide theoretical basis for achieving the ultimate goal of global quantum security communication. Keywords:continuous variable/ quantum key distribution/ discrete modulation/ quantum catalysis
由(15)式可知, 量子催化的成功概率${P_{\rm{d}}}$是与密钥率安全性边界(${K_0} = 0$)密切相关. 若${P_{\rm{d}}}=0$, 则${K_0} = 0$. 若$0 < {P_{\rm{d}}} \leqslant 1$, 则成功概率${P_{\rm{d}}}$不会影响密钥率的安全性边界. 此外, 根据(14)式, 图3给出了不同的调制方差V = 1.2, 1.3, 1.4, 1.5下量子催化的成功概率${P_{\rm{d}}}$随透射率T的变化. 显然, 在固定透射率T下, ${P_{\rm{d}}}$随着调制方差V减小而增大. 当给定调制方差V时, ${P_{\rm{d}}}$随着透射率T的增加而增大. 这意味着零光子催化易于实现, 极大地促进了输入-输出量子态之间的转换, 从而避免通信双方量子信息的丢失. 图 3 对于不同的调制方差V下量子催化的成功概率${P_{\rm{d}}}$随透射率T的变化(图中从上往下的虚线分别表示V = 1.2, 1.3, 1.4, 1.5) Figure3. Success probability of implementing such a zero-photon catalysis as a function of the transmittance T for several different V. The dashed lines from bottom to top correspond to V = 1.2, 1.3, 1.4, 1.5, respectively.
3.性能分析与讨论一般而言, 量子密钥分发协议的性能评估有3个重要指标: 安全密钥率(secret key rate)、最大安全传输距离(transmission distance)以及最大可容忍过噪声 (tolerable excess noise). 本文基于以上3个指标对所提的协议进行性能分析和讨论. 由图2可知, 在离散调制协议下, 调制方差V需控制在$V \in \left[ {1, \;\;1.5} \right)$范围内该方案才能与高斯调制的连续变量量子密钥分发协议等价, 这使得求解Holevo信息问题大大简化. 在信道损耗为0.2 dB/km下, 假设$\beta = 0.95, \xi = 0.005$, 对于不同的调制方差V = 1.2, 1.3, 1.4, 当优化透射率T, 图4(a)为基于量子催化的四态协议在不同距离下的安全码率. 图中黑色线为原始四态协议(original four-state modulation protocol, original protocol). 当调制方差取某些值(如1.3, 1.4), 本文的零光子催化四态协议(zero-photon catalysis-based four-state modulation protocol, ZPC protocol)能够在最大安全传输距离及安全密钥率性能方面优于原始四态协议. 这是源于零光子催化实际是一种无噪衰减过程, 而无噪声衰减已在参考文献[24]证实了可以提升量子密钥分发系统的性能. 另一方面, 通过优化量子催化引入的透射率T来调控和获取最优调制方差, 使得进一步提高量子密钥分发协议的性能. 同时, 这里也给出了透射率T在不同距离下的曲线图, 如图4(b)所示. 值得注意的是, 当T = 1时, 不存在任何量子催化效果. 正是如此, 一方面, 导致了在短距离安全通信下, 本方案与原始方案的性能保持一致. 另一方面, 如图4(a)所示, 这也使得对于V = 1.2的所提方案(蓝色虚线)与原始方案(黑色实线)的性能曲线重合. 这意味当调制方差低于某个值, 量子催化不能用来提高离散调制协议的性能. 此外, 从图4(a)可知, 对于原始四态协议(黑色线)而言, 调制方差的减小, 可以提高安全传输距离. 有趣的是, 量子催化的引入可进一步提升原始四态协议的安全传输距离. 图 4 离散调制量子密钥分发系统的性能比较 (a) 固定参数$\beta = 0.95, \xi = 0.005$下, 当优化透射率T时, 密钥率在不同调制方差随传输距离的变化; (b) 对应 (a) 情况下, 透射率T随传输距离的变化 Figure4. Comparison of the performances between the original protocol and the ZPC-based four-state modulation protocol: (a) At a fixed $\beta = 0.95, \xi = 0.005$, the secret key rate as a function of the transmission distance with different V = 1.2, 1.3, 1.4, when optimized over the transmittance T; (b) the transmittance T as a function of the transmission distance corresponding to panel (a).
此外, 可容忍过噪声是影响性能的另一项重要指标. 为了清晰理解可容忍过噪声对性能的影响, 这里适当选取固定参数$\beta = 0.95, V = 1.3$, 当优化透射率T时, 对于不同可容忍过噪声$\xi =$ 0.002, 0.005, 0.008, 图5(a)显示了安全密钥率随传输距离的变化. 显然, 可容忍噪声过噪声越低, 量子密钥分发协议的性能越好. 此外, 仿真结果表明, 量子催化可以用来有效地改善量子密钥分发的性能. 尤其对于较小的可容忍过噪声, 改善效果比较明显. 这正如图5(a)所示, 对于可容忍过噪声取0.002(点划线)和0.005 (划线)时, 零光子催化四态协议在传输距离和安全密钥率方面都能优越于原始方案. 而对于$\xi = 0.008$的情况, 性能改善不明显. 这是因为透射率T随可容忍过噪声的增加而在安全传输距离上缩短(如图5(b)所示). 同时, 这也暗含着可容忍过噪声的增加可以抑制量子催化的效果. 图 5 离散调制量子密钥分发系统的性能比较 (a) 固定参数$\beta = 0.95, V = 1.3$下, 当优化透射率T时, 密钥率在不同可容忍过噪声随传输距离的变化; (b) 对应 (a) 情况下, 透射率T随传输距离的变化曲线 Figure5. Comparison of the performances between the original protocol and the ZPC-based four-state modulation protocol: (a) At a fixed $\beta = 0.95, V = 1.3$, the secret key rate as a function of the transmission distance with different $\xi = 0.002, 0.005, 0.008$, when optimized over the transmittance T; (b) the transmittance T as a function of the transmission distance corresponding to panel (a).
为了进一步研究协商效率对量子密钥分发性能的影响, 这里假定$V = 1.3, \xi = 0.005$, 对于优化透射率T的情况下, 图6(a)表示不同的协商效率$\beta =0.90, 0.95, 1.0$的安全密钥率随传输距离的变化. 显然地, 协商效率越高, 则量子密钥分发的性能表现越好. 特别是, 对于更为实际的协商效率为0.90(点划线)时, 采用量子催化操作能够提升原始方案的传输距离约至210 km, 密钥率为10–8 bits/pulse. 图6(b)给出对应量子催化四态调制协议的透射率T随传输距离的变化. 此外, 比较图6(a)与图5(a)可以看出, 可容忍过噪声对量子密钥分发的性能影响程度要大于协商效率的情况. 图 6 离散调制量子密钥分发系统的性能比较 (a) 固定参数$V = 1.3, \xi = 0.005$下, 当优化透射率T时, 密钥率在不同协商效率随传输距离的变化; (b) 对应 (a) 情况下, 透射率T随传输距离的变化曲线 Figure6. Comparison of the performances between the original protocol and the ZPC-based four-state modulation protocol: (a) At a fixed $V = 1.3, \xi = 0.005$, the secret key rate as a function of the transmission distance with different $\beta = 0.90, 0.95, 1.0$, when optimized over the transmittance T; (b) the transmittance T as a function of the transmission distance corresponding to panel (a).
可容忍过噪声是影响量子密钥分发性能的关键因素. 为了看清量子催化能否提升最大可容忍过噪声, 对于不同的协商效率$\beta =0.90, 0.95, 1.0$, 图7给出了最大可容忍过噪声随传输距离的变化. 本方案的性能提升随着协商效率的降低而呈现得更为明显. 例如, 对于给定可容忍过噪声为$\xi = 0.003$, 原始协议的传输距离可达到240 km附近; 而相同参数下, 本方案的传输距离大约达到320 km. 这些研究结果表明, 零光子催化可提升离散调制协议的可容忍过噪声. 此外, 需要注意的是, 图中优化透射率T指的是在可取范围$T \in \left[ {0, 1} \right]$内找到某个透射率使得密钥率最大(图4—图6)或者密钥率为零(图7). 图 7 在固定参数$V = 1.3$下, 当优化透射率T时, 可容忍过噪声在不同协商效率随传输距离的变化 Figure7. At a fixed $V = 1.3$, the tolerable excess noise between the original protocol and the ZPC-based four-state modulation protocol as a function of a transmission distance for several different $\beta = 0.90, 0.95, 1.0$, when optimized over T.