摘要
本文分析了网络流量数据的特性,针对传统预测算法在预测网络流量时的缺陷提出了一种基于相关分析的相关局域最小二乘支持向量机(LSSVM)预测算法.算法在对训练数据重构相空间后,利用相关分析同时从距离相关和时间相关的训练样本中选择最优的训练子集,结合自适应参数优化的LSSVM预测模型对小尺度网络流量进行预测.通过选用实际情况下的网络流量数据对算法进行测试验证,结果显示本文所提算法不仅优于传统的全局预测算法,同时也优于各种改进的局域预测算法.算法不仅在预测精度上取得大幅的性能提升,同时能够通过留一交叉验证法在预测之前就完成预测模型和训练子集的优化.
Real-time monitoring and forecasting technology for network tra?c has played an important role in network man-agement. Effective network tra?c prediction could analyze and solve problems before overload occurs, which significantly improves network availability. In this paper, after the vulnerability of traditional nonlinear prediction method in fore-casting modeling is analyzed, the relevant local (RL) forecast which is based on correlation analysis and the parameter optimization method based on pattern search (PS) is introduced. Using the correlation analysis, the optimal training subset is chosen from time-and distance-correlated training samples. On this basis, the prediction model is established by LSSVM. Finally network tra?c dataset collected from wired campus networks is studied for our experiments. And the results show that the relevant local LSSVM prediction method whose training set and parameters have been auto-matically optimized can effectively predict the small scale tra?c measurement data, and RL-LSSVM tra?c forecasting algorithm exhibits significantly good prediction accuracy for the data set compared with previous algorithm.
出处
《物理学报》
SCIE
EI
CAS
CSCD
北大核心
2014年第13期49-58,共10页
Acta Physica Sinica
基金
国防科技预研项目(批准号:208010201)资助的课题~~
关键词
网络流量预测
混沌时间序列预测
最小二乘支持向量机
局域预测
network traffic prediction
chaos time series forecasting
least squares support vector machine
local prediction