多模光纤网络异常入侵信号提纯方法
|
摘要
多模光纤网络由于模式色散大,使其带宽较低,在多用户间连通使用易受到异常信号入侵攻击,导致其安全性较差。为提高异常入侵信号的检测准确率,增强光纤网络安全性,提出一种经验模态分解方法结合深度学习的多模光纤网络异常入侵信号提纯方法。针对光纤网络容易发生流量拥塞的问题,进行异常流量,提高网络安全性,提出基于深度学习的光纤网络异常流量检测方法。构建光纤网络的信号传输结构模型,采用非线性时间序列重构方法进行光纤网络异常入侵信号的特征提取,检测异常信号的时频谱特征量,结合经验模态分解方法实现光纤网络异常入侵信号特征信息分离,采用深度学习算法进行异常入侵信号特征分解的自适应迭代,提高异常入侵信号检测的收敛性,将特征分离后的入侵信号输入到降噪滤波器中,进行信号提纯处理,完成网络异常入侵信号检测。仿真结果表明,采用该方法进行光纤网络异常入侵信号检测的抗干扰能力较强,准确检测概率较高,收敛性较好,从而提高了检测效率,改善了多模光纤网络的安全性。
Because of the large dispersion of the mode,the bandwidth of the multimode fiber network is low,and it is vulnerable to the intrusion attack of the abnormal signal in the connection between the multi-users,which leads to the poor security of the network. In order to improve the detection accuracy of abnormal intrusion signals and enhance the security of optical fiber networks,an empirical mode decomposition method combined with depth learning is proposed to purify abnormal intrusion signals from multi-mode optical networks. Aiming at the problem that the fiber network is prone to traffic congestion,this paper proposes an anomaly traffic detection method based on depth learning to improve the network security. The signal transmission structure model of fiber optic network is constructed,and the nonlinear time series reconstruction method is adopted. The feature extraction of abnormal intrusion signal in optical fiber network,the detection of time spectrum characteristic quantity of abnormal signal,and the separation of characteristic information of abnormal intrusion signal of optical fiber network with empirical mode decomposition method are realized. The adaptive iteration of feature decomposition of abnormal intrusion signal is carried out by using depth learning algorithm to improve the convergence of abnormal intrusion signal detection. The intrusion signal after feature separation is input into the denoising filter for signal purification. Complete network anomaly intrusion signal detection. The simulation results show that the proposed method has strong anti-interference ability,higher detection probability and better convergence. The measurement efficiency improves the security of multimode fiber network.
Because of the large dispersion of the mode,the bandwidth of the multimode fiber network is low,and it is vulnerable to the intrusion attack of the abnormal signal in the connection between the multi-users,which leads to the poor security of the network. In order to improve the detection accuracy of abnormal intrusion signals and enhance the security of optical fiber networks,an empirical mode decomposition method combined with depth learning is proposed to purify abnormal intrusion signals from multi-mode optical networks. Aiming at the problem that the fiber network is prone to traffic congestion,this paper proposes an anomaly traffic detection method based on depth learning to improve the network security. The signal transmission structure model of fiber optic network is constructed,and the nonlinear time series reconstruction method is adopted. The feature extraction of abnormal intrusion signal in optical fiber network,the detection of time spectrum characteristic quantity of abnormal signal,and the separation of characteristic information of abnormal intrusion signal of optical fiber network with empirical mode decomposition method are realized. The adaptive iteration of feature decomposition of abnormal intrusion signal is carried out by using depth learning algorithm to improve the convergence of abnormal intrusion signal detection. The intrusion signal after feature separation is input into the denoising filter for signal purification. Complete network anomaly intrusion signal detection. The simulation results show that the proposed method has strong anti-interference ability,higher detection probability and better convergence. The measurement efficiency improves the security of multimode fiber network.
引文
[1]刘学君,袁碧贤,卓思超,等.基于MSP430的多模光纤短距离数采通信系统[J].电子设计工程,2017,25(6):96-99.
[2]陆兴华,陈平华.基于定量递归联合熵特征重构的缓冲区流量预测算法[J].计算机科学,2015,42(4):68-71.
[3]陈健,黄青青,张倩武,等.基于光子灯笼的正交频分/模分复用IM-DD多模光纤传输系统[J].光学学报,2018,38(6):102-108.
[4]王龙,杨承志,吴宏超,等.基于FPGA的数字基带多模雷达信号源设计[J].电子技术应用,2016,42(8):87-90.
[5]李影,谭中伟,孙剑,等.阶跃型多模光纤的选择性模式激励[J].光电技术应用,2016,31(1):11-15.
[6]李威,顾海林,黄兴.网络被入侵后的信号检测系统设计与优化[J].现代电子技术,2017,40(3):58-61.
[7] NGUYEN T C,SHEN W,LUO Z,et al. Novel Data Integrity Verification Schemes in Cloud Storage[M]. Switzerland:Springer International Publishing,2015:115-125.
[8] ELDEMERDASH Y A,DOBRE O A,and LIAO B J. Blind identification of SM and Alamouti STBC-OFDM signals[J].IEEE Transactions on Wireless Communications,2015,14(2):972-982.
[9]尚朝轩,王品,韩壮志,彭刚.基于类决策树分类的特征层融合识别算法[J].控制与决策,2016,31(06):1009-1014.
[10]孙三山,汪帅,樊自甫.软件定义网络架构下基于流调度代价的数据中心网络拥塞控制路由算法[J].计算机应用,2016,36(7):1784-1788.
[11] HUANG Lei,ZHANG Jing,XU Xu,et al. Robust adaptive beamforming with a novel interference-plus-noise covariance matrix reconstruction method[J]. IEEE Transactions on Signal Processing,2015,63(7):1643-1650.
[12]马俊涛,高梅国,董健.基于稀疏迭代协方差估计的缺失数据谱分析及时域重建方法[J].电子与信息学报,2016,38(6):1431-1437.
[13]顾艳林.大数据驱动下网络入侵信号提取检测仿真[J].计算机仿真,2017,34(9):370-373.
[14] JIANG X,HARISHAN K,THAMARASA R,et al. Integrated track initialization and maintenance in heavy clutter using probabilistic data association[J]. Signal Processing,2014,94:241-250.
[15] VLACHOS E,LALOS A S,and BERBERIDIS K. Stochastic gradient pursuit for adaptive equalization of sparse multipath channels[J]. IEEE Journal on Emerging and Selected Topics in Circuits and Systems,2012,2(3):413-423.
[16] HANSEN T L,BADIU M,FLEURY B H,et al. A sparse Bayesian learning algorithm with dictionary parameter estimation[C]//Sensor Array and Multichannel Signal Processing Workshop,A Coru,Spain,2014:385-388.
[17]蒋立辉,刘杰生,熊兴隆,等.光纤周界入侵信号特征提取与识别方法的研究[J].激光与红外,2017,47(7):906-913.
[18]陈霄,肖起榕,闫平,等.高功率光纤激光信号耦合器[J].红外与毫米波学报,2016,35(1):15-20.
[2]陆兴华,陈平华.基于定量递归联合熵特征重构的缓冲区流量预测算法[J].计算机科学,2015,42(4):68-71.
[3]陈健,黄青青,张倩武,等.基于光子灯笼的正交频分/模分复用IM-DD多模光纤传输系统[J].光学学报,2018,38(6):102-108.
[4]王龙,杨承志,吴宏超,等.基于FPGA的数字基带多模雷达信号源设计[J].电子技术应用,2016,42(8):87-90.
[5]李影,谭中伟,孙剑,等.阶跃型多模光纤的选择性模式激励[J].光电技术应用,2016,31(1):11-15.
[6]李威,顾海林,黄兴.网络被入侵后的信号检测系统设计与优化[J].现代电子技术,2017,40(3):58-61.
[7] NGUYEN T C,SHEN W,LUO Z,et al. Novel Data Integrity Verification Schemes in Cloud Storage[M]. Switzerland:Springer International Publishing,2015:115-125.
[8] ELDEMERDASH Y A,DOBRE O A,and LIAO B J. Blind identification of SM and Alamouti STBC-OFDM signals[J].IEEE Transactions on Wireless Communications,2015,14(2):972-982.
[9]尚朝轩,王品,韩壮志,彭刚.基于类决策树分类的特征层融合识别算法[J].控制与决策,2016,31(06):1009-1014.
[10]孙三山,汪帅,樊自甫.软件定义网络架构下基于流调度代价的数据中心网络拥塞控制路由算法[J].计算机应用,2016,36(7):1784-1788.
[11] HUANG Lei,ZHANG Jing,XU Xu,et al. Robust adaptive beamforming with a novel interference-plus-noise covariance matrix reconstruction method[J]. IEEE Transactions on Signal Processing,2015,63(7):1643-1650.
[12]马俊涛,高梅国,董健.基于稀疏迭代协方差估计的缺失数据谱分析及时域重建方法[J].电子与信息学报,2016,38(6):1431-1437.
[13]顾艳林.大数据驱动下网络入侵信号提取检测仿真[J].计算机仿真,2017,34(9):370-373.
[14] JIANG X,HARISHAN K,THAMARASA R,et al. Integrated track initialization and maintenance in heavy clutter using probabilistic data association[J]. Signal Processing,2014,94:241-250.
[15] VLACHOS E,LALOS A S,and BERBERIDIS K. Stochastic gradient pursuit for adaptive equalization of sparse multipath channels[J]. IEEE Journal on Emerging and Selected Topics in Circuits and Systems,2012,2(3):413-423.
[16] HANSEN T L,BADIU M,FLEURY B H,et al. A sparse Bayesian learning algorithm with dictionary parameter estimation[C]//Sensor Array and Multichannel Signal Processing Workshop,A Coru,Spain,2014:385-388.
[17]蒋立辉,刘杰生,熊兴隆,等.光纤周界入侵信号特征提取与识别方法的研究[J].激光与红外,2017,47(7):906-913.
[18]陈霄,肖起榕,闫平,等.高功率光纤激光信号耦合器[J].红外与毫米波学报,2016,35(1):15-20.