机器学习在量子通信资源优化配置中的应用

陈以鹏; 刘靖阳; 朱佳莉; 方伟; 王琴

doi:10.7498/aps.71.20220871

摘要

在未来量子通信网络的大规模应用中, 如何根据当前用户实际情况实现资源优化配置, 比如选择最优量子密钥分发协议(quantum key distribution, QKD)和最优系统参数等, 是实现网络应用的一个重要考察指标. 传统的QKD最优协议选择以及参数优化配置方法, 大多是通过局部搜索算法来实现. 该方法需要花费大量的计算资源和时间. 为此, 本文提出了将机器学习算法应用到QKD资源优化配置之中, 通过回归机器学习的方式来同时进行不同情境下的最优协议选择以及最优协议的参数优化配置. 此外, 将包括随机森林(random forest, RF)、最近邻(k-nearest neighbor, KNN)、逻辑回归(logistic regression)等在内的多种回归机器学习模型进行对比分析. 数据仿真结果显示, 基于机器学习的新方案与基于局部搜索算法的传统方案相比, 在资源损耗方面实现了质的跨越, 而且RF在多个回归评估指标上都取得了最佳的效果. 此外, 通过残差分析, 发现以RF回归模型为代表的机器学习方案在最优协议选择以及参数优化配置方面具有很好的环境鲁棒性. 因此, 本工作将对未来量子通信网络实际应用起到重要的推进作用.

关键词:

Abstract

In the application of quantum communication networks, it is an important task to realize the optimal allocation of resources according to the current situation. For example, We need to select the optimal quantum key distribution (QKD) protocol and parameters. Traditionally, the most commonly implemented method is the local search algorithm (LSA), which costs a lot of resources. Here in this work, we propose a machine learning based scheme, in which the regression machine learning is used to simultaneously select the optimal protocol and corresponding parameters. In addition, we make comparisons among a few machine learning models including random forest (RF), K-nearest neighbor (KNN) and logistic regression. Simulation results show that the new scheme takes much less time than the LSA scheme, and the RF achieves the best performance. In addition, through the RF residual analysis, we find that the machine learning scheme has good robustness. In conclusion, this work may play an important role in promoting the practical application of quantum communication networks.

Keywords:

作者及机构信息

1.
南京邮电大学, 量子信息技术研究所, 南京　210003

2.
南京邮电大学, 宽带无线通信与传感网教育部重点实验室, 南京　210003

通信作者: 王琴, qinw@njupt.edu.cn

基金项目: 国家重点研究发展计划(批准号: 2018YFA0306400, 2017YFA0304100) 、国家自然科学基金 (批准号: 12074194) 、江苏省自然科学基金前沿技术(批准号: BK20192001)和江苏省研究生科研创新计划(批准号: KYCX20_0726)资助的课题

Authors and contacts

1.
Institute of Quantum Information and Technology, Nanjing University of Posts and Telecommunications, Nanjing 210003, China

2.
Key Laboratory of Broadband Wireless Communication and Sensor Network of Ministry of Education, Nanjing University of Posts and Telecommunications, Nanjing 210003, China

Corresponding author: Wang Qin, qinw@njupt.edu.cn

Funds: Project supported by the National Key R&D Program of China (Grant Nos. 2018 YFA0306400, 2017 YFA0304100), the National Natural Science Foundation of China (Grant No. 12074194), the Leading-edge technology program of Jiangsu Natural Science Foundation (Grant No. BK20192001), and Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX20_0726).

文章全文 : translate this paragraph

参考文献

[1]	Bennett C H, Brassard G 1984 Proceedings of IEEE International Conference on Computers, System and Signal Processing (Bangalore: IEEE) p175
[2]	Busch P, Heinonen T, Lathi P 2007 Phys. Rep. 452 155 Google Scholar
[3]	Wootters W K, Zurek W H 1982 Nature. 299 299 Google Scholar
[4]	Hwang W Y 2003 Phys. Rev. Lett. 91 057901 Google Scholar
[5]	Wang X B 2005 Phys. Rev. Lett. 94 230503 Google Scholar
[6]	Lo H K, Ma X F, Chen K 2005 Phys. Rev. Lett. 94 230504 Google Scholar
[7]	Makarov V, Hjelme D R 2005 J. Mod. Optic. 52 691 Google Scholar
[8]	Qi B, Fung C H F, Lo H K, Ma X F 2007 Quantum. Inf. Comput. 7 73
[9]	Lamas L A, Qin L, Gerhardt I, Makarov V, Kurtsiefer C 2009 New. J. Phys. 11 065003 Google Scholar
[10]	Lydersen L, Wiechers C, Wittmann C, Elser D, Skaar J 2010 Nat. Photonics. 4 686 Google Scholar
[11]	Lo H, Curty M, Qi B 2012 Phys. Rev. Lett. 108 130503 Google Scholar
[12]	Braunstein S L, Pirandola S 2012 Phys. Rev. Lett. 108 130502 Google Scholar
[13]	Wang X B. 2013 Phys. Rev. A 87 012320 Google Scholar
[14]	Rubenok A, Slater J A, Chan P, Lucio M I, Tittel W 2013 Phys. Rev. Lett. 111 130501 Google Scholar
[15]	Tang Z Y, Liao Z F, Xu F H, Qi B, Qian L, Lo H K 2014 Phys. Rev. Lett. 112 190503 Google Scholar
[16]	Liu Y, Chen T Y, Wang L J, Liang H, Shentu G L, Wang J, Cui K, Yin H L, Liu N L, Li L, Ma X F, Pelc J S, Fejer M M, Peng C Z, Zhang Q, Pan J W 2013 Phys. Rev. Lett. 111 130502 Google Scholar
[17]	Zhou X Y, Ding H J, Zhang C H, Wang Q 2020 Opt. Lett. 45 4176 Google Scholar
[18]	Liu J Y, Zhou X Y Wang Q 2021 Phys. Rev. A. 103 022602 Google Scholar
[19]	Lucamarini M, Yuan Z L, Dynes J F, Shields A J 2018 Nature 557 400 Google Scholar
[20]	Takeoka M, Guha S. 2014 Nat. Commun. 5 5235 Google Scholar
[21]	Pirandola S, Laurenza R, Ottaviani C 2017 Nat. Commun. 8 15043 Google Scholar
[22]	Wang X B, Yu Z W, Hu X L 2018 Phys. Rev. A. 98 062323 Google Scholar
[23]	Pittaluga M, Minder M, Lucamarini M, Sanzaro M, Woodward R I, Li M J, Yuan Z L, Shields A J 2021 Nat. Photonics. 15 530 Google Scholar
[24]	Wang S, Yin Z Q, Chen W, He D Y, Song X T, Li H W, Zhang L J, Zhou Z, Guo G C, Han Z F 2022 Nat. Photonics. 16 154 Google Scholar
[25]	Ren Z A, Chen Y P, Liu J Y, Ding H J, Wang Q 2021 IEEE Commun. Lett. 25 3 Google Scholar
[26]	Fan-Yuan G J, Lu F Y, Wang S, Yin Z Q, He D Y, Zhou Z, Teng J, Chen W, Guo G C, Han Z F 2021 Photon. Res. 9 1881 Google Scholar
[27]	Ding H J, Liu J Y, Zhang C M, Wang Q 2020 Quantum. Inf. Comput. 19 2548 Google Scholar
[28]	Xu F, Xu H, Lo H K. 2014 Phys. Rev. A. 89 052333 Google Scholar
[29]	Liu J Y, Ding H J, Zhang C M, Xie S P, Wang Q 2019 Phys. Rev. Appl. 12 014059 Google Scholar
[30]	Yang M, Ren C L, Ma Y C, Xiao Y, Ye X J, Song L L, Xun J S, Yung M H, Li C F, Guo G C 2019 Phys. Rev. Lett. 123 190401 Google Scholar
[31]	Zhou Y H, Yu Z W, Wang X B. 2016 Phys. Rev. A. 93 042324 Google Scholar
[32]	Zhang C H, Zhang C M, Wang Q. 2019 Opt. Lett. 44 1468 Google Scholar
[33]	Breiman L 2001 J. Clin. Microbiol. 45 5 Google Scholar
[34]	Cover T M, Hart P E 1967 IEEE Trans. Inf. Theory 13 21 Google Scholar
[35]	Cox D R 1958 J. R. Stat. Soc. B 20 215 Google Scholar

施引文献

图 1 随机森林回归模型的算法框架.

Fig. 1. The algorithm framework of random forest regression model.

下载: 全尺寸图片幻灯片

图 2 系统参数对随机森林回归模型的重要性.

Fig. 2. Importance of system parameters to RF regression model.

下载: 全尺寸图片幻灯片

图 3 随机森林回归模型的残差图

Fig. 3. Residual diagram of RF regression model.

下载: 全尺寸图片幻灯片

图 4 随机森林回归模型的混淆矩阵

Fig. 4. Residual diagram of RF regression model.

下载: 全尺寸图片幻灯片

表 1 系统参数的特征范围

Table 1. Characteristic range of system parameters.

Y₀ e_d $ \eta $ N L/km

10^–10—10^–5 0.00—0.06 0.1—0.9 10⁶—10¹⁶ 1—600

下载: 导出CSV

表 2 不同回归模型的评估对比

Table 2. Evaluation and comparison of different regression models.

RF KNN LR

MAE 0.002 0.012 0.038
MSE 0.016 0.049 0.131
$ {\mathit{R}}^{2} $ 0.978 0.795 0.397
Accuracy 0.977 0.787 0.365

下载: 导出CSV

表 3 时间资源损耗记录表

Table 3. Time resource wastage table.

机器学习方案传统方案

Model RF KNN LR LSA
Time 1.23 s 2.95 s 5.43 s 24 h以上

下载: 导出CSV

[1]	Bennett C H, Brassard G 1984 Proceedings of IEEE International Conference on Computers, System and Signal Processing (Bangalore: IEEE) p175
[2]	Busch P, Heinonen T, Lathi P 2007 Phys. Rep. 452 155 Google Scholar
[3]	Wootters W K, Zurek W H 1982 Nature. 299 299 Google Scholar
[4]	Hwang W Y 2003 Phys. Rev. Lett. 91 057901 Google Scholar
[5]	Wang X B 2005 Phys. Rev. Lett. 94 230503 Google Scholar
[6]	Lo H K, Ma X F, Chen K 2005 Phys. Rev. Lett. 94 230504 Google Scholar
[7]	Makarov V, Hjelme D R 2005 J. Mod. Optic. 52 691 Google Scholar
[8]	Qi B, Fung C H F, Lo H K, Ma X F 2007 Quantum. Inf. Comput. 7 73
[9]	Lamas L A, Qin L, Gerhardt I, Makarov V, Kurtsiefer C 2009 New. J. Phys. 11 065003 Google Scholar
[10]	Lydersen L, Wiechers C, Wittmann C, Elser D, Skaar J 2010 Nat. Photonics. 4 686 Google Scholar
[11]	Lo H, Curty M, Qi B 2012 Phys. Rev. Lett. 108 130503 Google Scholar
[12]	Braunstein S L, Pirandola S 2012 Phys. Rev. Lett. 108 130502 Google Scholar
[13]	Wang X B. 2013 Phys. Rev. A 87 012320 Google Scholar
[14]	Rubenok A, Slater J A, Chan P, Lucio M I, Tittel W 2013 Phys. Rev. Lett. 111 130501 Google Scholar
[15]	Tang Z Y, Liao Z F, Xu F H, Qi B, Qian L, Lo H K 2014 Phys. Rev. Lett. 112 190503 Google Scholar
[16]	Liu Y, Chen T Y, Wang L J, Liang H, Shentu G L, Wang J, Cui K, Yin H L, Liu N L, Li L, Ma X F, Pelc J S, Fejer M M, Peng C Z, Zhang Q, Pan J W 2013 Phys. Rev. Lett. 111 130502 Google Scholar
[17]	Zhou X Y, Ding H J, Zhang C H, Wang Q 2020 Opt. Lett. 45 4176 Google Scholar
[18]	Liu J Y, Zhou X Y Wang Q 2021 Phys. Rev. A. 103 022602 Google Scholar
[19]	Lucamarini M, Yuan Z L, Dynes J F, Shields A J 2018 Nature 557 400 Google Scholar
[20]	Takeoka M, Guha S. 2014 Nat. Commun. 5 5235 Google Scholar
[21]	Pirandola S, Laurenza R, Ottaviani C 2017 Nat. Commun. 8 15043 Google Scholar
[22]	Wang X B, Yu Z W, Hu X L 2018 Phys. Rev. A. 98 062323 Google Scholar
[23]	Pittaluga M, Minder M, Lucamarini M, Sanzaro M, Woodward R I, Li M J, Yuan Z L, Shields A J 2021 Nat. Photonics. 15 530 Google Scholar
[24]	Wang S, Yin Z Q, Chen W, He D Y, Song X T, Li H W, Zhang L J, Zhou Z, Guo G C, Han Z F 2022 Nat. Photonics. 16 154 Google Scholar
[25]	Ren Z A, Chen Y P, Liu J Y, Ding H J, Wang Q 2021 IEEE Commun. Lett. 25 3 Google Scholar
[26]	Fan-Yuan G J, Lu F Y, Wang S, Yin Z Q, He D Y, Zhou Z, Teng J, Chen W, Guo G C, Han Z F 2021 Photon. Res. 9 1881 Google Scholar
[27]	Ding H J, Liu J Y, Zhang C M, Wang Q 2020 Quantum. Inf. Comput. 19 2548 Google Scholar
[28]	Xu F, Xu H, Lo H K. 2014 Phys. Rev. A. 89 052333 Google Scholar
[29]	Liu J Y, Ding H J, Zhang C M, Xie S P, Wang Q 2019 Phys. Rev. Appl. 12 014059 Google Scholar
[30]	Yang M, Ren C L, Ma Y C, Xiao Y, Ye X J, Song L L, Xun J S, Yung M H, Li C F, Guo G C 2019 Phys. Rev. Lett. 123 190401 Google Scholar
[31]	Zhou Y H, Yu Z W, Wang X B. 2016 Phys. Rev. A. 93 042324 Google Scholar
[32]	Zhang C H, Zhang C M, Wang Q. 2019 Opt. Lett. 44 1468 Google Scholar
[33]	Breiman L 2001 J. Clin. Microbiol. 45 5 Google Scholar
[34]	Cover T M, Hart P E 1967 IEEE Trans. Inf. Theory 13 21 Google Scholar
[35]	Cox D R 1958 J. R. Stat. Soc. B 20 215 Google 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]	刘天乐, 徐枭, 付博伟, 徐佳歆, 刘靖阳, 周星宇, 王琴. 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. 物理学报, 2023, 72(11): 110304. doi: 10.7498/aps.72.20230160
[3]	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
[4]	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
[5]	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
[6]	聂敏, 刘广腾, 杨光, 裴昌幸. 基于最少中继节点约束的量子VoIP路由优化策略. 物理学报, 2016, 65(12): 120302. doi: 10.7498/aps.65.120302
[7]	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
[8]	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
[9]	权东晓, 裴昌幸, 朱畅华, 刘丹. 一种新的预报单光子源诱骗态量子密钥分发方案. 物理学报, 2008, 57(9): 5600-5604. doi: 10.7498/aps.57.5600
[10]	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
[11]	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
[12]	郑力明, 王发强, 刘颂豪. 光纤色散与损耗对光量子密钥分发系统的影响. 物理学报, 2007, 56(4): 2180-2183. doi: 10.7498/aps.56.2180
[13]	赵峰, 路轶群, 王发强, 陈霞, 李明明, 郭邦红, 廖常俊, 刘颂豪. 基于微弱相干脉冲稳定差分相位量子密钥分发. 物理学报, 2007, 56(4): 2175-2179. doi: 10.7498/aps.56.2175
[14]	何广强, 易智, 朱俊, 曾贵华. 基于双模压缩态的量子密钥分发方案. 物理学报, 2007, 56(11): 6427-6433. doi: 10.7498/aps.56.6427
[15]	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
[16]	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
[17]	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
[18]	李明明, 王发强, 路轶群, 赵峰, 陈霞, 梁瑞生, 刘颂豪. 高稳定的差分相位编码量子密钥分发系统. 物理学报, 2006, 55(9): 4642-4646. doi: 10.7498/aps.55.4642
[19]	马海强, 李亚玲, 赵环, 吴令安. 基于双偏振分束器的量子密钥分发系统. 物理学报, 2005, 54(11): 5014-5017. doi: 10.7498/aps.54.5014
[20]	吴光, 周春源, 陈修亮, 韩晓红, 曾和平. 长距离长期稳定的量子密钥分发系统. 物理学报, 2005, 54(8): 3622-3626. doi: 10.7498/aps.54.3622

计量

文章访问数: 3439
PDF下载量: 86
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

搜索

留言板

机器学习在量子通信资源优化配置中的应用