搜索

x

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于回归决策树的测量设备无关型量子密钥分发参数优化

刘天乐 徐枭 付博伟 徐佳歆 刘靖阳 周星宇 王琴

引用本文:
Citation:

基于回归决策树的测量设备无关型量子密钥分发参数优化

刘天乐, 徐枭, 付博伟, 徐佳歆, 刘靖阳, 周星宇, 王琴

Regression-decision-tree based parameter optimization of measurement-device-independent quantum key distribution

Liu Tian-Le, Xu Xiao, Fu Bo-Wei, Xu Jia-Xin, Liu Jing-Yang, Zhou Xing-Yu, Wang Qin
PDF
HTML
导出引用
  • 量子密钥分发(quantum key distribution, QKD)结合一次一密的加密方式, 可以实现无条件安全的量子通信. 双场(twin-field, TF) QKD和测量设备无关(measurement-device-independent, MDI) QKD具有较高的安全性, 同时适合构建以测量端为中心的网络, 具有广阔的应用前景. 但在实际应用过程中, 参数配置对QKD性能有着极大影响, 而实际场景中存在着用户数量大、位置距离中心站点非对称、并且用户大部分处在实时移动中的特点. 面对上述实时的参数配置需求, 传统的参数优化方式将无法满足. 本文提出将监督机器学习算法应用于QKD参数优化配置中, 通过机器学习模型预测不同场景下TF和MDI两种常用协议的最优参数. 将神经网络、最近邻、随机森林、梯度提升决策树和分类回归决策树(classification and regression tree, CART)等监督学习模型进行对比, 结果显示CART模型在$ {R^2} $等回归评估指标上均有最优表现. 在随机划分训练组、验证组情况下, 预测参数的密钥率与最优密钥率比值的均值在0.995以上; 在“超精度”和“超范围”两种极限情况下, 该均值仍能维持在0.988左右, 且在残差分析中具有较好的环境鲁棒性, 展现出较好的性能. 此外, 基于CART的新方案相较于传统方案在计算实时性表现上有很大提升, 将单次预测时间缩短至微秒量级, 很好地满足了通信方在移动状态下的实时通信需求.
    The parameter configuration of quantum key distribution (QKD) has a great effect on the communication effect, and in the practical application of the QKD network in the future, it is necessary to quickly realize the parameter configuration optimization of the asymmetric channel measurement-device-independent QKD according to the communication state, so as to ensure the good communication effect of the mobile users, which is an inevitable requirement for real-time quantum communication. Aiming at the problem that the traditional QKD parameter optimization configuration scheme cannot guarantee real-time, in this paper we propose to apply the supervised machine learning algorithm to the QKD parameter optimization configuration, and predict the optimal parameters of TF-QKD and MDI-QKD under different conditions through the machine learning model. First, we delineate the range of system parameters and evenly spaced (linear or logarithmic) values through experimental experience, and then use the traditional local search algorithm (LSA) to obtain the optimal parameters and take them as the optimal parameters in this work. Finally, we train various machine learning models based on the above data and compare their performances. We compare the supervised regression learning models such as neural network, K-nearest neighbors, random forest, gradient tree boosting and classification and regression tree (CART), and the results show that the CART decision tree model has the best performance in the regression evaluation index, and the average value of the key rate (of the prediction parameters) and the optimal key rate ratio is about 0.995, which can meet the communication needs in the actual environment. At the same time, the CART decision tree model shows good environmental robustness in the residual analysis of asymmetric QKD protocol. In addition, compared with the traditional scheme, the new scheme based on CART decision tree greatly improves the real-time performance of computing, shortening the single prediction time of the optimal parameters of different environments to the microsecond level, which well meets the real-time communication needs of the communicator in the movable state. This work mainly focuses on the parameter optimization of discrete variable QKD (DV-QKD). In recent years, the continuous variable QKD (CV-QKD) has developed also rapidly. At the end of the paper, we briefly introduce academic attempts of applying machine learning to the parameter optimization of CV-QKD system, and discuss the applicability of the scheme in CV-QKD system.
      通信作者: 周星宇, xyz@njupt.edu.cn
    • 基金项目: 国家重点研发计划 (批准号: 2018YFA0306400)、国家自然科学基金 (批准号: 12074194, 62101285, 62201276)和江苏省自然科学基金前沿引领技术 (批准号: BK20192001)资助的课题.
      Corresponding author: Zhou Xing-Yu, xyz@njupt.edu.cn
    • Funds: Project supported by the National Key R&D Program of China (Grant No. 2018YFA0306400), the National Natural Science Foundation of China (Grant Nos. 12074194, 62101285, 62201276), and the Leading-edge Technology Program of Jiangsu Natural Science Foundation, China (Grant No. BK20192001).
    [1]

    Gisin N, Thew R 2007 Nat. Photonics. 1 165Google Scholar

    [2]

    Scarani V, Bechmann P H, Cerf N J, Dusek M, Lutkenhaus N, Peev M 2009 Rev. Mod. Phys. 81 1301Google Scholar

    [3]

    Wootters W K, Zurek W H 1982 Nature 299 802Google Scholar

    [4]

    Busch P, Heinonen T, Lathi P 2007 Phys. Rep. 452 155Google Scholar

    [5]

    Deutsch D, Ekert A, Jozsa R, Macchiavello C, Popescu S, Sanpera A 1996 Phys. Rev. Lett. 77 2818Google Scholar

    [6]

    Bennett C H, Brassard G 2014 Theoret. Comput. Sci. 560 7Google Scholar

    [7]

    杨林轩, 苏志锟 2022 中国高新科技 11 82Google Scholar

    Yang L X, Su Z K 2022 China High and New Technol. 11 82Google Scholar

    [8]

    Lütkenhaus N 2000 Phys. Rev. A 61 052304Google Scholar

    [9]

    Acin A, Gisin N, Scarani V 2004 Phys. Rev. A 69 012309Google Scholar

    [10]

    Lydersen L, Wiechers C, Wittmann C, Elser D, Skaar J, Makarov V 2010 Nat. Photonics. 4 686Google Scholar

    [11]

    Lo H K, Curty M, Qi B 2012 Phys. Rev. Lett. 108 130503Google Scholar

    [12]

    Braunstein S L, Pirandola S 2012 Phys. Rev. Lett. 108 130502Google Scholar

    [13]

    Lucamarini M, Yuan Z L, Dynes J F, Shields A J 2018 Nature. 557 400Google Scholar

    [14]

    Takeoka M, Guha S, Wilde M M 2014 Nat. Commun. 5 5235Google Scholar

    [15]

    Wang X B, Yu Z W, Hu X L 2018 Phys. Rev. A 98 062323Google Scholar

    [16]

    Ma X, Zeng P, Zhou H 2018 Phys. Rev. X 8 031043

    [17]

    王华, 赵永利 2019 通信学报 40 168Google Scholar

    Wang H, Zhao Y L 2019 J. Commun. 40 168Google Scholar

    [18]

    Hughes R J, Morgan G L, Peterson C G 2000 J. Mod. Opt. 47 533

    [19]

    Ren Z A, Chen Y P, Liu J Y, Ding H J, Wang Q 2020 IEEE Commun. Lett. 25 940

    [20]

    Ding H J, Liu J Y, Zhang C M, Wang Q 2020 Quant. Inform. Proces. 19 1Google Scholar

    [21]

    Xu F, Xu H, Lo H K 2014 Phys. Rev. A 89 052333Google Scholar

    [22]

    Liu W, Huang P, Peng J, Fan J, Zeng G 2018 Phys. Rev. A 97 022316Google Scholar

    [23]

    Wang W, Lo H K 2019 Phys. Rev. A 100 062334Google Scholar

    [24]

    Lu F Y, Yin Z Q, Wang C, Cui C H, Teng J, Wang S, Chen W, Huang W, Xu B J, Guo G C, Han Z F 2019 JOSA B 36 B92Google Scholar

    [25]

    陈以鹏, 刘靖阳, 朱佳莉, 方伟, 王琴 2022 物理学报 71 220301Google Scholar

    Chen Y P, Liu J Y, Zhu J L, Fang W, Wang Q 2022 Acta Phys. Sin. 71 220301Google Scholar

    [26]

    王琴, 陈以鹏 2020 南京邮电大学学报 40 141

    Wang Q, Chen Y P 2020 J. Nanjing University of Posts and Telecommun. 40 141

    [27]

    Cao Y, Li Y H, Yang K X, et al. 2020 Phys. Rev. Lett. 125 260503Google Scholar

    [28]

    Zhou X Y, Zhang C H, Zhang C M, Wang Q 2019 Phys. Rev. A 99 062316Google Scholar

    [29]

    Wang W, Xu F, Lo H K 2019 Phys. Rev. X 9 041012

    [30]

    Quinlan J R 1986 Mach. Learn. 1 81

    [31]

    Rumelhart D E, Hinton G E, Williams R J 1986 Nature. 323 533Google Scholar

    [32]

    Gordon A D, Breiman L, Friedman J H, Olshen R A, Stone C J 1984 Biometrics. 40 874Google Scholar

    [33]

    申媛媛, 邬锦雯, 刘鑫东 2020 科技管理研究 40 91

    Shen Y Y, Wu T W, Liu X D 2020 Sci. Technol. Manage. Res. 40 91

    [34]

    刘勇洪, 牛铮, 王长耀 2005 遥感学报 9 405Google Scholar

    Liu Y H, Niu Z, Wang C Y 2005 J. Remote Sens. 9 405Google Scholar

    [35]

    王辉, 张文杰, 刘杰, 陈林烽, 李泽南 2022 中国民航大学学报 40 35Google Scholar

    Wang H, Zhang W J, Liu J, Chen L F, Li Z N 2022 J. Civil Aviation University of China 40 35Google Scholar

    [36]

    刘玉茹, 赵成萍, 臧军, 宁芊, 周新志 2017 计算机应用 37 57

    Liu Y R, Zhao C P, Zang J, Ning Q, Zhou X Z 2017 Comput. Appl. 37 57

    [37]

    S. Pirandola, Andersen U L, Banchi L, et al. 2020 Adv. Opt. Photonics 12 1012Google Scholar

    [38]

    Huang D, Liu S, Zhang L 2021 Photonics 8 511Google Scholar

    [39]

    Liu Z P, Zhou M G, Liu W B, Li C L, Gu J, Yin H L, Chen Z B 2022 Opt. Express 30 15024Google Scholar

    [40]

    Luo H, Wang Y J, Ye W, Zhong H, Mao Y Y, Guo Y 2022 Phys. B 31 020306

    [41]

    Zhou M G, Liu Z P, Liu W B, Li C L, Bai J L, Xue Y R, Fu Y, Yin H L, Chen Z B 2022 Sci. Rep. 12 8879Google Scholar

  • 图 1  MDI-QKD系统示意图

    Fig. 1.  Schematic diagram of MDI-QKD system.

    图 2  不同类别监督学习算法$ {R^2} $比较

    Fig. 2.  Comparison of R2 of supervised learning algorithms in different categories.

    图 3  标准情况决策树模型残差图

    Fig. 3.  Decision tree model residual plot for standard cases.

    图 4  标准情况神经网络模型残差图

    Fig. 4.  Neural networks model residual plot for standard cases.

    图 5  标准情况KNN模型残差图

    Fig. 5.  KNN model residual plot for standard cases.

    图 6  CART构建过程

    Fig. 6.  CART construction process.

    图 7  TF-QKD模型比例柱状图

    Fig. 7.  Model scale histogram of TF-QKD.

    图 8  MDI-QKD模型比例柱状图

    Fig. 8.  Model scale histogram of MDI-QKD.

    图 9  回归决策树模型比例柱状图

    Fig. 9.  Regression decision tree model scale histogram.

    图 10  RF模型比例柱状图

    Fig. 10.  Random forest model scale histogram.

    图 11  超精度情况决策树模型残差图

    Fig. 11.  Decision tree model residual plot for super-precision cases.

    图 12  超精度情况RF模型残差图

    Fig. 12.  RF model residual plot for super-precision cases.

    表 1  系统参数范围

    Table 1.  System parameter range.

    参数$ L $/km$ \Delta L $/km$ N $$ \eta $$ {Y_0} $$ {e_{\text{d}}} $
    范围0—3000, 25, 50, 75, 100109—10140.1—0.910–11—10–60.01—0.10
    下载: 导出CSV

    表 2  不同方案时间消耗对比

    Table 2.  Comparison of time loss between different schemes.

    协议决策树/s神经网络/sKNN/s传统/h
    TF-QKD0.7130.8251.509163
    MDI-QKD0.6080.7101.301116
    下载: 导出CSV

    表 3  标准情况下模型结果评估

    Table 3.  Evaluation of the results under standard conditions.

    协议R2MAE/10–3MSE/10–5
    TF-QKD0.99163.428.05
    MDI-QKD0.99930.371.70
    下载: 导出CSV

    表 4  不同使用场景模型结果评估

    Table 4.  Evaluation of the results of different usage scenarios.

    模型$ {R^2} $MAE/10–2MSE/10–4
    超范围决策树0.95291.648.75
    RF0.94941.689.12
    梯度提升0.85792.7640.0
    超精度决策树0.96591.143.52
    RF0.96541.153.54
    梯度提升0.91541.948.74
    注: 粗体数据为该指标最好的结果.
    下载: 导出CSV

    表 5  不同方案时间消耗对比

    Table 5.  Comparison of time loss between different schemes.

    协议RF/s决策树/s梯度提升树/s传统/h
    TF-QKD1.4260.7136.748163
    MDI-QKD1.2210.6085.631116
    下载: 导出CSV
  • [1]

    Gisin N, Thew R 2007 Nat. Photonics. 1 165Google Scholar

    [2]

    Scarani V, Bechmann P H, Cerf N J, Dusek M, Lutkenhaus N, Peev M 2009 Rev. Mod. Phys. 81 1301Google Scholar

    [3]

    Wootters W K, Zurek W H 1982 Nature 299 802Google Scholar

    [4]

    Busch P, Heinonen T, Lathi P 2007 Phys. Rep. 452 155Google Scholar

    [5]

    Deutsch D, Ekert A, Jozsa R, Macchiavello C, Popescu S, Sanpera A 1996 Phys. Rev. Lett. 77 2818Google Scholar

    [6]

    Bennett C H, Brassard G 2014 Theoret. Comput. Sci. 560 7Google Scholar

    [7]

    杨林轩, 苏志锟 2022 中国高新科技 11 82Google Scholar

    Yang L X, Su Z K 2022 China High and New Technol. 11 82Google Scholar

    [8]

    Lütkenhaus N 2000 Phys. Rev. A 61 052304Google Scholar

    [9]

    Acin A, Gisin N, Scarani V 2004 Phys. Rev. A 69 012309Google Scholar

    [10]

    Lydersen L, Wiechers C, Wittmann C, Elser D, Skaar J, Makarov V 2010 Nat. Photonics. 4 686Google Scholar

    [11]

    Lo H K, Curty M, Qi B 2012 Phys. Rev. Lett. 108 130503Google Scholar

    [12]

    Braunstein S L, Pirandola S 2012 Phys. Rev. Lett. 108 130502Google Scholar

    [13]

    Lucamarini M, Yuan Z L, Dynes J F, Shields A J 2018 Nature. 557 400Google Scholar

    [14]

    Takeoka M, Guha S, Wilde M M 2014 Nat. Commun. 5 5235Google Scholar

    [15]

    Wang X B, Yu Z W, Hu X L 2018 Phys. Rev. A 98 062323Google Scholar

    [16]

    Ma X, Zeng P, Zhou H 2018 Phys. Rev. X 8 031043

    [17]

    王华, 赵永利 2019 通信学报 40 168Google Scholar

    Wang H, Zhao Y L 2019 J. Commun. 40 168Google Scholar

    [18]

    Hughes R J, Morgan G L, Peterson C G 2000 J. Mod. Opt. 47 533

    [19]

    Ren Z A, Chen Y P, Liu J Y, Ding H J, Wang Q 2020 IEEE Commun. Lett. 25 940

    [20]

    Ding H J, Liu J Y, Zhang C M, Wang Q 2020 Quant. Inform. Proces. 19 1Google Scholar

    [21]

    Xu F, Xu H, Lo H K 2014 Phys. Rev. A 89 052333Google Scholar

    [22]

    Liu W, Huang P, Peng J, Fan J, Zeng G 2018 Phys. Rev. A 97 022316Google Scholar

    [23]

    Wang W, Lo H K 2019 Phys. Rev. A 100 062334Google Scholar

    [24]

    Lu F Y, Yin Z Q, Wang C, Cui C H, Teng J, Wang S, Chen W, Huang W, Xu B J, Guo G C, Han Z F 2019 JOSA B 36 B92Google Scholar

    [25]

    陈以鹏, 刘靖阳, 朱佳莉, 方伟, 王琴 2022 物理学报 71 220301Google Scholar

    Chen Y P, Liu J Y, Zhu J L, Fang W, Wang Q 2022 Acta Phys. Sin. 71 220301Google Scholar

    [26]

    王琴, 陈以鹏 2020 南京邮电大学学报 40 141

    Wang Q, Chen Y P 2020 J. Nanjing University of Posts and Telecommun. 40 141

    [27]

    Cao Y, Li Y H, Yang K X, et al. 2020 Phys. Rev. Lett. 125 260503Google Scholar

    [28]

    Zhou X Y, Zhang C H, Zhang C M, Wang Q 2019 Phys. Rev. A 99 062316Google Scholar

    [29]

    Wang W, Xu F, Lo H K 2019 Phys. Rev. X 9 041012

    [30]

    Quinlan J R 1986 Mach. Learn. 1 81

    [31]

    Rumelhart D E, Hinton G E, Williams R J 1986 Nature. 323 533Google Scholar

    [32]

    Gordon A D, Breiman L, Friedman J H, Olshen R A, Stone C J 1984 Biometrics. 40 874Google Scholar

    [33]

    申媛媛, 邬锦雯, 刘鑫东 2020 科技管理研究 40 91

    Shen Y Y, Wu T W, Liu X D 2020 Sci. Technol. Manage. Res. 40 91

    [34]

    刘勇洪, 牛铮, 王长耀 2005 遥感学报 9 405Google Scholar

    Liu Y H, Niu Z, Wang C Y 2005 J. Remote Sens. 9 405Google Scholar

    [35]

    王辉, 张文杰, 刘杰, 陈林烽, 李泽南 2022 中国民航大学学报 40 35Google Scholar

    Wang H, Zhang W J, Liu J, Chen L F, Li Z N 2022 J. Civil Aviation University of China 40 35Google Scholar

    [36]

    刘玉茹, 赵成萍, 臧军, 宁芊, 周新志 2017 计算机应用 37 57

    Liu Y R, Zhao C P, Zang J, Ning Q, Zhou X Z 2017 Comput. Appl. 37 57

    [37]

    S. Pirandola, Andersen U L, Banchi L, et al. 2020 Adv. Opt. Photonics 12 1012Google Scholar

    [38]

    Huang D, Liu S, Zhang L 2021 Photonics 8 511Google Scholar

    [39]

    Liu Z P, Zhou M G, Liu W B, Li C L, Gu J, Yin H L, Chen Z B 2022 Opt. Express 30 15024Google Scholar

    [40]

    Luo H, Wang Y J, Ye W, Zhong H, Mao Y Y, Guo Y 2022 Phys. B 31 020306

    [41]

    Zhou M G, Liu Z P, Liu W B, Li C L, Bai J L, Xue Y R, Fu Y, Yin H L, Chen Z B 2022 Sci. Rep. 12 8879Google Scholar

  • [1] 周江平, 周媛媛, 周学军. 非对称信道相位匹配量子密钥分发. 物理学报, 2023, 72(14): 140302. doi: 10.7498/aps.72.20230652
    [2] 朱佳莉, 曹原, 张春辉, 王琴. 实用化量子密钥分发光网络中的资源优化配置. 物理学报, 2023, 72(2): 020301. doi: 10.7498/aps.72.20221661
    [3] 周阳, 马啸, 周星宇, 张春辉, 王琴. 实用化态制备误差容忍参考系无关量子密钥分发协议. 物理学报, 2023, 72(24): 240301. doi: 10.7498/aps.72.20231144
    [4] 周江平, 周媛媛, 周学军. 改进的测量设备无关协议参数优化方法. 物理学报, 2023, 72(12): 120303. doi: 10.7498/aps.72.20230179
    [5] 孟杰, 徐乐辰, 张成峻, 张春辉, 王琴. 标记单光子源在量子密钥分发中的应用. 物理学报, 2022, 71(17): 170304. doi: 10.7498/aps.71.20220344
    [6] 陈以鹏, 刘靖阳, 朱佳莉, 方伟, 王琴. 机器学习在量子通信资源优化配置中的应用. 物理学报, 2022, 71(22): 220301. doi: 10.7498/aps.71.20220871
    [7] 叶炜, 郭迎, 夏莹, 钟海, 张欢, 丁建枝, 胡利云. 基于量子催化的离散调制连续变量量子密钥分发. 物理学报, 2020, 69(6): 060301. doi: 10.7498/aps.69.20191689
    [8] 杜聪, 王金东, 秦晓娟, 魏正军, 於亚飞, 张智明. 基于混合编码的测量设备无关量子密钥分发的简单协议. 物理学报, 2020, 69(19): 190301. doi: 10.7498/aps.69.20200162
    [9] 谷文苑, 赵尚弘, 东晨, 王星宇, 杨鼎. 参考系波动下的参考系无关测量设备无关量子密钥分发协议. 物理学报, 2019, 68(24): 240301. doi: 10.7498/aps.68.20191364
    [10] 吴承峰, 杜亚男, 王金东, 魏正军, 秦晓娟, 赵峰, 张智明. 弱相干光源测量设备无关量子密钥分发系统的性能优化分析. 物理学报, 2016, 65(10): 100302. doi: 10.7498/aps.65.100302
    [11] 杜亚男, 解文钟, 金璇, 王金东, 魏正军, 秦晓娟, 赵峰, 张智明. 基于弱相干光源测量设备无关量子密钥分发系统的误码率分析. 物理学报, 2015, 64(11): 110301. doi: 10.7498/aps.64.110301
    [12] 贾晓洁, 艾斌, 许欣翔, 杨江海, 邓幼俊, 沈辉. 选择性发射极晶体硅太阳电池的二维器件模拟及性能优化. 物理学报, 2014, 63(6): 068801. doi: 10.7498/aps.63.068801
    [13] 周飞, 雍海林, 李东东, 印娟, 任继刚, 彭承志. 基于不同介质间量子密钥分发的研究. 物理学报, 2014, 63(14): 140303. doi: 10.7498/aps.63.140303
    [14] 李红祺, 郭维栋, 孙国栋, 张耀存. 条件非线性最优扰动方法在陆面过程模式参数优化中的扩展应用初探. 物理学报, 2011, 60(1): 019201. doi: 10.7498/aps.60.019201
    [15] 胡华鹏, 张 静, 王金东, 黄宇娴, 路轶群, 刘颂豪, 路 巍. 双协议量子密钥分发系统实验研究. 物理学报, 2008, 57(9): 5605-5611. doi: 10.7498/aps.57.5605
    [16] 张 静, 王发强, 赵 峰, 路轶群, 刘颂豪. 时间和相位混合编码的量子密钥分发方案. 物理学报, 2008, 57(8): 4941-4946. doi: 10.7498/aps.57.4941
    [17] 张新陆, 王月珠, 李 立, 崔金辉, 鞠有伦. 端面抽运Tm,Ho∶YLF连续激光器的参数优化与实验研究. 物理学报, 2008, 57(6): 3519-3524. doi: 10.7498/aps.57.3519
    [18] 冯发勇, 张 强. 基于超纠缠交换的量子密钥分发. 物理学报, 2007, 56(4): 1924-1927. doi: 10.7498/aps.56.1924
    [19] 陈 霞, 王发强, 路轶群, 赵 峰, 李明明, 米景隆, 梁瑞生, 刘颂豪. 运行双协议相位调制的量子密钥分发系统. 物理学报, 2007, 56(11): 6434-6440. doi: 10.7498/aps.56.6434
    [20] 陈 杰, 黎 遥, 吴 光, 曾和平. 偏振稳定控制下的量子密钥分发. 物理学报, 2007, 56(9): 5243-5247. doi: 10.7498/aps.56.5243
计量
  • 文章访问数:  2776
  • PDF下载量:  61
  • 被引次数: 0
出版历程
  • 收稿日期:  2023-02-08
  • 修回日期:  2023-03-17
  • 上网日期:  2023-03-28
  • 刊出日期:  2023-06-05

/

返回文章
返回