基于时延光子储备池计算的混沌激光短期预测

刘奇; 李璞; 开超; 胡春强; 蔡强; 张建国; 徐兵杰

doi:10.7498/aps.70.20210355

摘要

提出并证明了一种利用时延光子储备池计算短期预测混沌激光的时间序列. 具体来说, 建立基于光反馈和光注入半导体激光器的储备池结构, 通过选择合适的系统参数, 时延光子储备池计算可以有效地预测混沌激光约2 ns的动态轨迹. 此外, 研究了系统参数对预测结果的影响, 包括掩模类型、虚拟节点数、训练数据长度、输入增益、反馈强度、注入强度、岭参数和泄漏率. 作为一种具有全光实现潜力的机器学习方法, 时延光子储备池具有结构简单、训练成本低、易于硬件实现等优点.

关键词:

Abstract

Prediction of chaotic laser has a wide prospect of applications, such as retrieving lost data, providing assists for data analysis, testing data encryption security in cryptography based on chaotic synchronization of lasers. We propose and demonstrate a new method of using time delayed photonic reservoir computing (RC) to forecast the continuous dynamical evolution of chaotic laser from previous measurements. Specifically, the time delayed photonic RC based on semiconductor laser with optical injection and feedback structure is established as a prediction system. Chaotic laser, as input signal, is generated by semiconductor laser with external disturbance.The time delayed photonic RC used in this stage is a novel implementation, which consists of three parts: the input layer, the reservoir and the output layer. In the input layer, the chaos laser from the semiconductor with an optical feedback needs to preprocess and multiply by a mask signal. The reservoir is the master-slave configuration consisting of a response laser with the optical feedback and light injection. In the feedback loop, there are N virtual nodes at each interval θ with a delay time of τ (N = τ/θ). The reservoir performs the mapping of the input signal onto a high-dimensional state space. In the output layer, the output of the reservoir is a linear combination of the reservoir state and the output weight. The output weight is optimized by minimizing the mean-square error between target value and output value through using the ridge regression algorithm.The results demonstrate that time delayed photonic RC based on semiconductor laser can forecast the trajectory of chaotic laser in about 2 ns. Moreover, we also investigate the influence of critical parameters on prediction result, including the type of the mask, the quantity of the virtual nodes, the length of the training data, the input gain, the feedback strength, the injection strength, the ridge parameter and the leakage rate.The method used here in this work has many attractive advantages, such as simple configuration, low training cost and eminently suitable for hardware implementation. Although the prediction length is limited, the significant innovation using time delayed photonic RC based on semiconductor lasers as the prediction system of chaotic laser presents a new opportunity for further developing a technique for predicting chaotic laser.

Keywords:

作者及机构信息

1.
太原理工大学, 新型传感器与智能控制教育部重点实验室, 太原　030024

2.
广东工业大学信息工程学院, 广州　510006

3.
广东省光子学信息技术重点实验室, 广州　510006

4.
中国电子科技集团公司第三十研究所, 成都　610041

5.
西南通信研究所, 保密通信重点实验室, 成都　610041

通信作者: 李璞, lipu8603@126.com

基金项目: 国家自然科学基金(批准号: 61775158, 61805168, 61961136002, 61927811, U19A2076, 11904057)、国家密码局“十三五”国家发展基金(批准号: MMJJ20170127)、中国博士后科学基金(批准号: 2018M630283, 2019T120197)、山西省自然科学基金(批准号: 201901D211116)和山西省高等学校优秀青年学术带头人计划资助的课题

Authors and contacts

1.
Key Laboratory of Advanced Transducers and Intelligent Control System, Ministry of Education, Taiyuan University of Technology, Taiyuan 030024, China

2.
School of Information Engineering, Guangdong University of Technology, Guangdong 510006, China

3.
Guangdong Key Laboratory of Photonics Information Technology, Guangdong 510006, China

4.
No.30 Institute of China Electronic Technology Corporation, Chengdu 610041, China

5.
Science and Technology on Communication Laboratory, Institute of Southwestern Communication, Chengdu 610041, China

Corresponding author: Li Pu, lipu8603@126.com

Funds: Project supported by the National Natural Science Foundation of China (Grant Nos. 61775158, 61805168, 61961136002, 61927811, U19A2076, 11904057), the National Cryptography Development Fund, China (Grant No. MMJJ20170127), the China Postdoctoral Science Foundation (Grant Nos. 2018M630283, 2019T120197), the Natural Science Foundation of Shanxi Province, China (Grant No. 201901D211116), and the Program for the Top Young Academic Leaders of High Learning Institutions of Shanxi Province, China

文章全文 : translate this paragraph

参考文献

[1]	Amil P, Soriano M C, Masoller C 2019 Chaos 29 113111 Google Scholar
[2]	Cunillera A, Soriano M C, Fischer I 2019 Chaos 29 113113 Google Scholar
[3]	Meng Q F, Peng Y H 2007 Phys. Lett. A 370 465 Google Scholar
[4]	Lin X W, Yang Z H, Song Y X 2009 Expert Syst. Appl. 36 7313 Google Scholar
[5]	Pathak J, Wikner A, Fussell R, Chandra S, Hunt B R, Girvan M, Ott E 2018 Chaos 28 041101 Google Scholar
[6]	Pathak J, Hunt B, Girvan M, Lu Z X, Ott E 2018 Phys. Rev. Lett. 120 024102 Google Scholar
[7]	Covas E, Benetos E 2019 Chaos 29 063111 Google Scholar
[8]	Appeltant L, Soriano M C, Van d S G, Danckaert J, Massar S, Dambre J, Schrauwen B, Mirasso C R, Fischer I 2011 Nat. Commun. 2 468 Google Scholar
[9]	Hicke K, Escalona-Moran M A, Brunner D, Soriano M C, Fischer I, Mirasso C R 2013 IEEE J. Sel. Top. Quantum Electron. 19 4 Google Scholar
[10]	Paquot Y, Duport F, Smerieri A, Dambre J, Schrauwen B, Haelterman M, Massar S 2012 Sci. Rep. 2 287 Google Scholar
[11]	Duport F, Schneider B, Smerieri A, Haelterman M, Massar S 2012 Opt. Express 20 A20 Google Scholar
[12]	Martinenghi R, Rybalko S, Jacquot M, Chembo Y K, Larger L 2012 Phys. Rev. Lett. 108 244101 Google Scholar
[13]	Larger L, Soriano M C, Brunner D, Appeltant L, Gutierrez J M, Pesquera L, Mirasso C R, Fischer I 2012 Opt. Express 20 3241 Google Scholar
[14]	Dejonckheere A, Duport F, Smerieri A, Fang L, Oudar J L, Haelterman M, Massar S 2014 Opt. Express 22 9 Google Scholar
[15]	Appeltant L, Sande G V D, Danckaert J, Fischer I 2014 Sci. Rep. 4 3629 Google Scholar
[16]	Nguimdo R M, Verschaffelt G, Danckaert J, Van der Sande G 2014 Opt. Express 22 8672 Google Scholar
[17]	Nguimdo R M, Verschaffelt G, Danckaert J, Van der Sande G 2015 IEEE Trans. Neural Networks Learn. Syst. 26 3301 Google Scholar
[18]	Vinckier Q, Duport F, Smerieri A, Vandoorne K, Bienstman P, Haelterman M, Massar S 2015 Optica 2 438 Google Scholar
[19]	Larger L, Baylón-Fuentes A, Martinenghi R, Udaltsov V S, Chembo Y K, Jacquot M 2017 Phys. Rev. X 7 011015 Google Scholar
[20]	Takano K, Sugano C, Inubushi M, Yoshimura K, Sunada S, Kanno K, Uchida A 2018 Opt. Express 26 29424 Google Scholar
[21]	Vatin J, Rontani D, M Sciamanna 2018 Opt. Lett. 43 4497 Google Scholar
[22]	Vatin J, Rontani D, M Sciamanna 2019 Opt. Express 27 018579 Google Scholar
[23]	Tan X S, Hou Y S, Wu Z M, Xia G Q 2019 Opt. Express 27 026082 Google Scholar
[24]	Guo X X, Xiang S Y, Zhang Y H, Lin L, Wen A J, Hao Y 2019 IEEE J. Sel. Top. Quantum Electron. 26 1700109 Google Scholar
[25]	Li P, Cai Q, Zhang J, Xu B, Wang Y 2019 Opt. Express 27 017859 Google Scholar
[26]	Nakayama J, Kanno K, Uchida A 2016 Opt. Express 24 8679 Google Scholar
[27]	Lang R, Kobayashi K 1980 IEEE J. Quantum Electron. 16 3 Google Scholar
[28]	Wieczorek S, Krauskopf B, Simpson T B, Lenstra D 2005 Phys. Rep. 416 1 Google Scholar
[29]	Hoerl A E, Kennard R W 1970 Technometrics 12 1 Google Scholar
[30]	Lukosevicius M, Jaeger H 2009 Comput. Sci. Rev. 3 127 Google Scholar

施引文献

图 1 基于半导体激光器的时延储备池计算系统示意图

Fig. 1. Schematic diagram of reservoir computing system based on semiconductor laser.

下载: 全尺寸图片幻灯片

图 2 (a) 激光器输出强度的分岔图; (b) 混沌激光目标信号及其预测值

Fig. 2. (a) Bifurcation diagram of the output intensity of the laser; (b) chaotic laser target signal together with the predicted values.

下载: 全尺寸图片幻灯片

图 3 (a) 在不同掩模类型和不同节点数的情况下, 预测长度(P_L)随训练长度(T_s)变化的趋势图; (b) 在使用混沌掩模信号且节点数为800的情况下, 预测长度(P_L)随输入增益(G_in)变化的趋势图; 虚线为拟合曲线

Fig. 3. (a) P_L as a function of the length of the training data (T_s) under different type of masks and the number of nodes; (b) P_L as a function of the input gain (G_in) under N = 800 with the chaos mask signal. The dotted lines represent the associated fitting curves.

下载: 全尺寸图片幻灯片

图 4 (a)预测系统在不同反馈强度(k_f)和注入强度(k_inj)的参数空间中P_L值的二维图; (b), (c) 在(k_inj, k_f) = (0.06, 0.18)的条件下, 无调制信号和有调制信号时R-laser的输出强度时序

Fig. 4. (a) Two-dimensional map of the P_L values of prediction system in the parameter space of the different feedback strength (k_f) and the injection strength (k_inj); (b), (c) temporal traces of the R-laser under (k_inj, k_f) = (0.06, 0.18) without and with modulated input data.

下载: 全尺寸图片幻灯片

图 5 (a) 预测长度(P_L)随泄漏率(δ)变化的趋势图; (b) 预测长度(P_L)随岭参数(λ)变化的趋势图; 虚线为拟合曲线

Fig. 5. (a) P_L as a function of the leakage rate (δ); (b) P_L as a function of the Ridge parameter (λ). The dotted lines represent the associated fitting curves.

下载: 全尺寸图片幻灯片

表 1 数值模拟中使用的激光器参数值

Table 1. Laser parameter values used in numerical simulations.

符号	参数	参考值
q/C	电子电荷量	1.6 × 10^–19
α	线宽增强因子	5.0
g/(m³·s^–1)	微分增益	1.414 × 10^–12
N₀/ m^-3	透明载流子密度	1.4 × 10²⁴
ε	增益饱和系数	5.0 × 10^–23
τ_p/ps	光子寿命	1.92
τ_s/ns	载流子寿命	2.04
τ_in/ps	内腔往返时间	7.38
k_f	反馈强度	0.18
k_inj	注入强度	0.06
∆ν/GHz	频率失谐	–10
τ/ns	外腔反馈延时	8 (N = 800)
θ/ns	节点间隔	0.01
I_d/I_th	D-laser的归一化偏置电流	1.2
I_r/I_th	R-laser的归一化偏置电流	1.25
G_in	输入增益	1.5
λ	岭参数	10^–6
δ	泄漏率	0.25

下载: 导出CSV

[1]	Amil P, Soriano M C, Masoller C 2019 Chaos 29 113111 Google Scholar
[2]	Cunillera A, Soriano M C, Fischer I 2019 Chaos 29 113113 Google Scholar
[3]	Meng Q F, Peng Y H 2007 Phys. Lett. A 370 465 Google Scholar
[4]	Lin X W, Yang Z H, Song Y X 2009 Expert Syst. Appl. 36 7313 Google Scholar
[5]	Pathak J, Wikner A, Fussell R, Chandra S, Hunt B R, Girvan M, Ott E 2018 Chaos 28 041101 Google Scholar
[6]	Pathak J, Hunt B, Girvan M, Lu Z X, Ott E 2018 Phys. Rev. Lett. 120 024102 Google Scholar
[7]	Covas E, Benetos E 2019 Chaos 29 063111 Google Scholar
[8]	Appeltant L, Soriano M C, Van d S G, Danckaert J, Massar S, Dambre J, Schrauwen B, Mirasso C R, Fischer I 2011 Nat. Commun. 2 468 Google Scholar
[9]	Hicke K, Escalona-Moran M A, Brunner D, Soriano M C, Fischer I, Mirasso C R 2013 IEEE J. Sel. Top. Quantum Electron. 19 4 Google Scholar
[10]	Paquot Y, Duport F, Smerieri A, Dambre J, Schrauwen B, Haelterman M, Massar S 2012 Sci. Rep. 2 287 Google Scholar
[11]	Duport F, Schneider B, Smerieri A, Haelterman M, Massar S 2012 Opt. Express 20 A20 Google Scholar
[12]	Martinenghi R, Rybalko S, Jacquot M, Chembo Y K, Larger L 2012 Phys. Rev. Lett. 108 244101 Google Scholar
[13]	Larger L, Soriano M C, Brunner D, Appeltant L, Gutierrez J M, Pesquera L, Mirasso C R, Fischer I 2012 Opt. Express 20 3241 Google Scholar
[14]	Dejonckheere A, Duport F, Smerieri A, Fang L, Oudar J L, Haelterman M, Massar S 2014 Opt. Express 22 9 Google Scholar
[15]	Appeltant L, Sande G V D, Danckaert J, Fischer I 2014 Sci. Rep. 4 3629 Google Scholar
[16]	Nguimdo R M, Verschaffelt G, Danckaert J, Van der Sande G 2014 Opt. Express 22 8672 Google Scholar
[17]	Nguimdo R M, Verschaffelt G, Danckaert J, Van der Sande G 2015 IEEE Trans. Neural Networks Learn. Syst. 26 3301 Google Scholar
[18]	Vinckier Q, Duport F, Smerieri A, Vandoorne K, Bienstman P, Haelterman M, Massar S 2015 Optica 2 438 Google Scholar
[19]	Larger L, Baylón-Fuentes A, Martinenghi R, Udaltsov V S, Chembo Y K, Jacquot M 2017 Phys. Rev. X 7 011015 Google Scholar
[20]	Takano K, Sugano C, Inubushi M, Yoshimura K, Sunada S, Kanno K, Uchida A 2018 Opt. Express 26 29424 Google Scholar
[21]	Vatin J, Rontani D, M Sciamanna 2018 Opt. Lett. 43 4497 Google Scholar
[22]	Vatin J, Rontani D, M Sciamanna 2019 Opt. Express 27 018579 Google Scholar
[23]	Tan X S, Hou Y S, Wu Z M, Xia G Q 2019 Opt. Express 27 026082 Google Scholar
[24]	Guo X X, Xiang S Y, Zhang Y H, Lin L, Wen A J, Hao Y 2019 IEEE J. Sel. Top. Quantum Electron. 26 1700109 Google Scholar
[25]	Li P, Cai Q, Zhang J, Xu B, Wang Y 2019 Opt. Express 27 017859 Google Scholar
[26]	Nakayama J, Kanno K, Uchida A 2016 Opt. Express 24 8679 Google Scholar
[27]	Lang R, Kobayashi K 1980 IEEE J. Quantum Electron. 16 3 Google Scholar
[28]	Wieczorek S, Krauskopf B, Simpson T B, Lenstra D 2005 Phys. Rep. 416 1 Google Scholar
[29]	Hoerl A E, Kennard R W 1970 Technometrics 12 1 Google Scholar
[30]	Lukosevicius M, Jaeger H 2009 Comput. Sci. Rev. 3 127 Google Scholar

[1]	宋睿, 刘雪梅, 王海滨, 吕皓, 宋晓艳. 机器学习辅助的WC-Co硬质合金硬度预测. 物理学报, 2024, 73(12): 126201. doi: 10.7498/aps.73.20240284
[2]	张旭, 丁进敏, 侯晨阳, 赵一鸣, 刘鸿维, 梁生. 基于机器学习的激光匀光整形方法. 物理学报, 2024, 73(16): 164205. doi: 10.7498/aps.73.20240747
[3]	张嘉晖. 蛋白质计算中的机器学习. 物理学报, 2024, 73(6): 069301. doi: 10.7498/aps.73.20231618
[4]	张桥, 谭薇, 宁勇祺, 聂国政, 蔡孟秋, 王俊年, 朱慧平, 赵宇清. 基于机器学习和第一性原理计算的Janus材料的预测. 物理学报, 2024, 73(23): 230201. doi: 10.7498/aps.73.20241278
[5]	杨章章, 刘丽, 万致涛, 付佳, 樊群超, 谢锋, 张燚, 马杰. 结合机器学习算法提高从头算方法对HF/HBr/H³⁵Cl/Na³⁵Cl振动能谱的预测性能. 物理学报, 2023, 72(7): 073101. doi: 10.7498/aps.72.20221953
[6]	罗启睿, 沈一凡, 罗孟波. 高分子塌缩相变和临界吸附相变的计算机模拟和机器学习. 物理学报, 2023, 72(24): 240502. doi: 10.7498/aps.72.20231058
[7]	张逸凡, 任卫, 王伟丽, 丁书剑, 李楠, 常亮, 周倩. 机器学习结合固溶强化模型预测高熵合金硬度. 物理学报, 2023, 72(18): 180701. doi: 10.7498/aps.72.20230646
[8]	黎威, 龙连春, 刘静毅, 杨洋. 基于机器学习的无机磁性材料磁性基态分类与磁矩预测. 物理学报, 2022, 71(6): 060202. doi: 10.7498/aps.71.20211625
[9]	赵彤, 谢文丽, 许俊伟, 贾志伟. 短内腔激光器对光子储备池计算的优化. 物理学报, 2022, 71(19): 194205. doi: 10.7498/aps.71.20220774
[10]	梅英, 谭冠政, 刘振焘, 武鹤. 基于大脑情感学习模型和自适应遗传算法的混沌时间序列预测. 物理学报, 2018, 67(8): 080502. doi: 10.7498/aps.67.20172104
[11]	沈力华, 陈吉红, 曾志刚, 金健. 基于鲁棒极端学习机的混沌时间序列建模预测. 物理学报, 2018, 67(3): 030501. doi: 10.7498/aps.67.20171887
[12]	李军, 李大超. 基于优化核极限学习机的风电功率时间序列预测. 物理学报, 2016, 65(13): 130501. doi: 10.7498/aps.65.130501
[13]	田中大, 李树江, 王艳红, 高宪文. 短期风速时间序列混沌特性分析及预测. 物理学报, 2015, 64(3): 030506. doi: 10.7498/aps.64.030506
[14]	王新迎, 韩敏. 多元混沌时间序列的多核极端学习机建模预测. 物理学报, 2015, 64(7): 070504. doi: 10.7498/aps.64.070504
[15]	李军, 张友鹏. 基于高斯过程的混沌时间序列单步与多步预测. 物理学报, 2011, 60(7): 070513. doi: 10.7498/aps.60.070513
[16]	王永生, 孙瑾, 王昌金, 范洪达. 变参数混沌时间序列的神经网络预测研究. 物理学报, 2008, 57(10): 6120-6131. doi: 10.7498/aps.57.6120
[17]	张军峰, 胡寿松. 基于一种新型聚类算法的RBF神经网络混沌时间序列预测. 物理学报, 2007, 56(2): 713-719. doi: 10.7498/aps.56.713
[18]	贺涛, 周正欧. 基于分形自仿射的混沌时间序列预测. 物理学报, 2007, 56(2): 693-700. doi: 10.7498/aps.56.693
[19]	叶美盈, 汪晓东, 张浩然. 基于在线最小二乘支持向量机回归的混沌时间序列预测. 物理学报, 2005, 54(6): 2568-2573. doi: 10.7498/aps.54.2568
[20]	李军, 刘君华. 一种新型广义RBF神经网络在混沌时间序列预测中的研究. 物理学报, 2005, 54(10): 4569-4577. doi: 10.7498/aps.54.4569

计量

文章访问数: 5944
PDF下载量: 169
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