• Chinese Journal of Quantum Electronics
  • Vol. 40, Issue 4, 546 (2023)
SUN Yishi and SUN Yi*
Author Affiliations
  • College of Communication and Information Technology, Xi'an University of Science and Technology, Xi'an 710054, China
  • show less
    DOI: 10.3969/j.issn.1007-5461.2023.04.014 Cite this Article
    Yishi SUN, Yi SUN. Parameter prediction of classical-quantum signals co-fiber transmission system based on BP neural network[J]. Chinese Journal of Quantum Electronics, 2023, 40(4): 546 Copy Citation Text show less
    Structure diagram of WDM-QKD system
    Fig. 1. Structure diagram of WDM-QKD system
    Comparison of the detection rates of SRS noise, FWM noise and out-band noise
    Fig. 2. Comparison of the detection rates of SRS noise, FWM noise and out-band noise
    The secure key transmission relationship under different decoy state methods with statistical fluctuations
    Fig. 3. The secure key transmission relationship under different decoy state methods with statistical fluctuations
    Training performance of neural network with different distribution of neurons. (a) Neuron setup (3, 2, 1); (b) Neuron setup (6, 4, 1); (c) Neuron setup (15, 10, 1); (d) Neuron setup (48, 24, 1)
    Fig. 4. Training performance of neural network with different distribution of neurons. (a) Neuron setup (3, 2, 1); (b) Neuron setup (6, 4, 1); (c) Neuron setup (15, 10, 1); (d) Neuron setup (48, 24, 1)
    Training performance of neural network with LM variable gradient algorithm
    Fig. 5. Training performance of neural network with LM variable gradient algorithm
    Training performance of neural network with Bayesian regularization algorithm
    Fig. 6. Training performance of neural network with Bayesian regularization algorithm
    Training performance of neural network with SCG algorithm
    Fig. 7. Training performance of neural network with SCG algorithm
    BP neural network model
    Fig. 8. BP neural network model
    Light source prediction results using BP neural network. (a) Imitative effect; (b) Training error
    Fig. 9. Light source prediction results using BP neural network. (a) Imitative effect; (b) Training error

    输入: 训练集D=(xk,yk)k=1m; 学习率 η

    输出: 连接权值或阈值确定的多层前馈神经网络

    1: functionBPD,η

    2: 在 (0, 1) 范围内随机初始化网络中的所有连接权值和阈值

    3: repeat

    4: for all xk,ykD do

    5: 计算当前样本输出 ŷk=fβj-θj

    6: 计算输出神经元梯度

    gj=-Ekŷkŷkβj=-ŷjk-yjkf'βj-θj=ŷjk1-ŷjkŷjk-yjk

    7: 计算隐层的神经元梯度项

    eh=-Ekbhbhah=-j=1lEkβjβjbhf'ah-γh=j=1lwhjgjf'ah-γh=bh1-bhj=1lwhjgj

    8: 更新权值

    Δwhj=ngjbh

    Δvih=ηehxi

    9: 更新阈值

    Δθj=-ηgj

    Δγh=-ηeh

    10: end for

    11: until 达到停止条件

    12: end function

    Table 1. Propagation algorithm of BP neural network
    ParameterValue
    I (波分复用器的插入损耗)/dB1.5
    αc (光纤衰减系数)/(dB·km-1)0.21
    ξ (波分复用器的隔离度)/dB80
    τD (单光子探测器门宽)/ps100
    ηD (单光子探测器探测效率)0.045
    h (普朗克常数)/(J·s)6.63×10-34
    c (光速)/(m·s-1)299792458
    tFc (滤波后带外噪声的透过率)0.01
    tF (滤波后带内噪声的透过率)0.95
    tBob (来自内部光学器件的透过率)0.9
    β (自发拉曼散射系数)/(km·nm)2×10-9
    Δλ (滤波器带宽)/pm35.23
    χ (光纤非线性系数)/(W·km)1.2
    Dijk (简并因子)3
    Dc (光纤色散系数)/(ps·nm-1·km-1)18
    dDc/dλ (光纤色散斜率)/(ps·nm-2·km-1)0.056
    ed (未对准引起的误码率)0.033
    Pd (探测器的暗记数率)3×10-6
    f(Eμ) (纠错系数)1.22
    Table 2. Parameters used for the numerical simulation of WDM-QKD
    Yishi SUN, Yi SUN. Parameter prediction of classical-quantum signals co-fiber transmission system based on BP neural network[J]. Chinese Journal of Quantum Electronics, 2023, 40(4): 546
    Download Citation