摘要
网络流量建模预测是网络管理和安全预警的基础。为了提高网络流量的预测精度,提出一种改进布谷鸟搜索算法优化支持向量机的网络流量预测模型(MCS-SVM)。首先将一维网络流量时间序列重构成多维时间序列;然后将支持向量机参数看作一个鸟巢位置,通过模拟布谷种群寄生繁衍机制找到最优参数;最后根据最优参数建立网络流量预测模型,并通过仿真实验对MCSSVM的性能进行测试。仿真结果表明,相对于参比模型,MCS-SVM提高了网络流量的预测精度,更加准确地刻画了网络流量复杂变化趋势,为具有混沌性网络流量预测提供了一种新的研究工具。
Network traffic modelling and prediction is the base of network management and safety early warning. In order to improve the accuracy of network traffic prediction, we present a network traffic prediction model which is based on optimising the support vector machine with the improved cuckoo search algorithm （MCS-SVM）. First, we reconstruct one-dimension time series of the network traffic to a multidimensional time series, then we regard the parameters of support vector machine as the bird nest, and find out optimal parameters through simulating the parasitism mechanism of cuckoo population, finally we build the network traffic prediction model according to the optimal parameters, and test the performance of MCS-SVM by simulation experiments. The simulation results show that compared with reference models, the proposed model improves the accuracy of network traffic prediction and can more precisely describe the complex variation trend of network traffic, it provides a new research tool for the prediction of network traffic with chaotic property.
出处
《计算机应用与软件》
CSCD
2015年第1期124-127,共4页
Computer Applications and Software
基金
广东省重大科技专项(2011A080401003
2011A080802008)
广州市科技计划项目(2010Z1-D00061)
关键词
网络流量预测
支持向量机
布谷鸟搜索算法
混沌理论
Network traffic prediction Support vector machine Cuckoo search algorithm Chaotic theory
