基于Kalman算法的光纤网络流量在线预测模型
|
摘要
为了准确地对光纤网络中的流量进行预测,需要对光纤网络流量预测方法进行研究。采用当前方法预测光纤网络中的流量时,得到的预测结果与实际结果不符,预测所用的时间较长,存在预测结果准确率低、误差率高、预测效率低的问题。在Kalman算法的基础上提出一种光纤网络流量在线预测模型,在预测光纤网络流量之前,对光纤网络中的流量进行多尺度分析,通过细节信号和逼近信号无损地表示光纤网络信号;分析光纤网络的历史流量数据,根据分析结果估计光纤网络中的流量,采用Kalman算法构建光纤网络流量在线预测模型,对估计结果进行校正,完成光纤网络流量的在线预测。实验结果表明,所提方法得到的光纤网络流量预测结果的准确率高、误差率低、预测效率高。
In order to accurately predict the traffic in the fiber network,the fiber network traffic forecasting method needs to be studied. When the current method is used to predict the traffic in the optical fiber network,the obtained prediction result does not match the actual result. The prediction takes a long time,and there is a problem that the accuracy of the prediction result is low,the error rate is high,and the prediction efficiency is low. Based on the Kalman algorithm,an on-line online traffic prediction model for fiber network is proposed. The traffic in the fiber network is analyzed before multi-scale prediction. The detail signal and the approximation signal are used to represent the fiber network signal losslessly. Historical traffic data,based on the analysis results to estimate the traffic in the fiber network,using Kalman algorithm to build a fiber network traffic online forecasting model,corrects the estimation results,and completes online forecasting of fiber network traffic. Experimental results show that the optical network traffic prediction results obtained by the proposed method have high accuracy,low error rate,and high prediction efficiency.
In order to accurately predict the traffic in the fiber network,the fiber network traffic forecasting method needs to be studied. When the current method is used to predict the traffic in the optical fiber network,the obtained prediction result does not match the actual result. The prediction takes a long time,and there is a problem that the accuracy of the prediction result is low,the error rate is high,and the prediction efficiency is low. Based on the Kalman algorithm,an on-line online traffic prediction model for fiber network is proposed. The traffic in the fiber network is analyzed before multi-scale prediction. The detail signal and the approximation signal are used to represent the fiber network signal losslessly. Historical traffic data,based on the analysis results to estimate the traffic in the fiber network,using Kalman algorithm to build a fiber network traffic online forecasting model,corrects the estimation results,and completes online forecasting of fiber network traffic. Experimental results show that the optical network traffic prediction results obtained by the proposed method have high accuracy,low error rate,and high prediction efficiency.
引文
[1]倪飞祥.基于改进小波-Elman神经网络算法的蜂窝网流量预测[J].电子设计工程,2017,25(3):171-175.
[2]孙诗晴,初凤红.基于优化神经网络算法的光纤布拉格光栅电流传感器的温度补偿[J].光学学报,2017,37(10):61-67.
[3]刘春桐,张正义,李洪才,等.基于FBG的液压系统中流量/压力/温度同时测量技术[J].光子学报,2016,45(11):10-16.
[4]刘强,张晓雨,刘懿莹,等.弯曲光纤布拉格光栅靶式流量传感器的研究[J].应用光学,2017,38(2):336-340.
[5]张林,杨立,寇蔚,等.基于表面测温等方差拟合的管内流量及温度识别[J].红外与激光工程,2018,47(1):111-118.
[6]义理林,李响.高速光接入网系统关键技术的研究[J].光通信研究,2017,43(6):21-27.
[7]邝斌,刘人榕,吴雅婷,等.基于流量预测的OFDMAPON动态带宽分配算法[J].光通信技术,2016,40(5)11-14.
[8]房亮,赵光恒.光纤通道流量控制方法及其性能分析[J].航天器工程,2015,24(2):51-57.
[9]崔秀国,刘翔,操时宜,等.光纤通信系统技术的发展、挑战与机遇[J].电信科学,2016,32(5):34-43.
[10]田鸿.强电磁干扰下光纤网络的数据传输系统改进设计[J].现代电子技术,2016,39(20):53-56.
[11]刘焕淋,李瑞艳,孔德谦,等.基于多目标遗传算法优化弹性光网络的多路径保护机制[J].电子与信息学报,2016,38(9):2261-2267.
[12]熊余,高玉鹏,王汝言.弹性光网络中负载均衡的优先共享通路保护[J].上海交通大学学报,2017,51(5):610-618.
[13]许鹏,张继栋.基于改进遗传算法的光纤网络异常数据检测[J].科技通报,2016,32(7):163-166.
[14]郭士荦,许江宁,李峰,等.优化BP神经网络的光纤陀螺温度漂移建模与补偿[J].中国惯性技术学报,2016,24(1):93-97.
[15]田泽,徐文龙,许恒,等.FC光纤通道技术研究综述[J].电子技术应用,2016,42(9):143-146.
[16]王芳,朱晗,李云鹏,等.错位光纤干涉激光谱结合BP神经网络的温度传感研究[J].光谱学与光谱分析,2016,36(11):3732-3736.
[17]李红,祝连庆,闫光,等.基于改性型UDP和CIGI的光纤光栅高速解调技术研究[J].仪器仪表学报,2016,37(1):129-135.
[18]张博,叶中付,徐旭,等.一种基于二维算法的新颖的多目标光纤光谱数据处理流程[J].天文学报,2016,57(1):91-101.
[2]孙诗晴,初凤红.基于优化神经网络算法的光纤布拉格光栅电流传感器的温度补偿[J].光学学报,2017,37(10):61-67.
[3]刘春桐,张正义,李洪才,等.基于FBG的液压系统中流量/压力/温度同时测量技术[J].光子学报,2016,45(11):10-16.
[4]刘强,张晓雨,刘懿莹,等.弯曲光纤布拉格光栅靶式流量传感器的研究[J].应用光学,2017,38(2):336-340.
[5]张林,杨立,寇蔚,等.基于表面测温等方差拟合的管内流量及温度识别[J].红外与激光工程,2018,47(1):111-118.
[6]义理林,李响.高速光接入网系统关键技术的研究[J].光通信研究,2017,43(6):21-27.
[7]邝斌,刘人榕,吴雅婷,等.基于流量预测的OFDMAPON动态带宽分配算法[J].光通信技术,2016,40(5)11-14.
[8]房亮,赵光恒.光纤通道流量控制方法及其性能分析[J].航天器工程,2015,24(2):51-57.
[9]崔秀国,刘翔,操时宜,等.光纤通信系统技术的发展、挑战与机遇[J].电信科学,2016,32(5):34-43.
[10]田鸿.强电磁干扰下光纤网络的数据传输系统改进设计[J].现代电子技术,2016,39(20):53-56.
[11]刘焕淋,李瑞艳,孔德谦,等.基于多目标遗传算法优化弹性光网络的多路径保护机制[J].电子与信息学报,2016,38(9):2261-2267.
[12]熊余,高玉鹏,王汝言.弹性光网络中负载均衡的优先共享通路保护[J].上海交通大学学报,2017,51(5):610-618.
[13]许鹏,张继栋.基于改进遗传算法的光纤网络异常数据检测[J].科技通报,2016,32(7):163-166.
[14]郭士荦,许江宁,李峰,等.优化BP神经网络的光纤陀螺温度漂移建模与补偿[J].中国惯性技术学报,2016,24(1):93-97.
[15]田泽,徐文龙,许恒,等.FC光纤通道技术研究综述[J].电子技术应用,2016,42(9):143-146.
[16]王芳,朱晗,李云鹏,等.错位光纤干涉激光谱结合BP神经网络的温度传感研究[J].光谱学与光谱分析,2016,36(11):3732-3736.
[17]李红,祝连庆,闫光,等.基于改性型UDP和CIGI的光纤光栅高速解调技术研究[J].仪器仪表学报,2016,37(1):129-135.
[18]张博,叶中付,徐旭,等.一种基于二维算法的新颖的多目标光纤光谱数据处理流程[J].天文学报,2016,57(1):91-101.