一种基于KPCA-LSSVM的可用带宽在线预测算法

发布时间：2018-05-04 02:19

本文选题：可用带宽 + 在线预测　；参考：《计算机应用与软件》2014年10期

【摘要】：针对目前端到端可用带宽预测方面研究工作较少的现状,提出一种基于核主成分分析KPCA(Kernel Principle Component Analysis)和最小二乘支持向量机LSSVM(Least Squares Support Vector Machine)的可用带宽在线预测算法ABOP。在采集网络状态样本数据并对其进行相空间重构的基础上,采用KPCA对数据进行降维降噪处理,最后基于LSSVM对可用带宽进行在线预测。为减小计算开销,提出一种递推计算的方法加快模型更新速度,并采用粒子群优化算法对模型参数进行多步更新,确保了在线预测的时效性。仿真表明,提出的ABOP算法具有较高的预测精度和较快的预测速度,能够满足可用带宽在线预测的要求。
[Abstract]:In view of the lack of research on end-to-end available bandwidth prediction, an on-line prediction algorithm for available bandwidth based on kernel principal component analysis (KPCA(Kernel Principle Component Analysis) and least squares support vector machine (LSSVM(Least Squares Support Vector Machine) is proposed. On the basis of collecting the network state sample data and reconstructing the phase space, KPCA is used to reduce the dimension of the data and the available bandwidth is predicted online based on LSSVM. In order to reduce the computational overhead, a recursive computing method is proposed to accelerate the updating speed of the model, and the particle swarm optimization algorithm is used to update the parameters of the model in order to ensure the timeliness of on-line prediction. Simulation results show that the proposed ABOP algorithm has higher prediction accuracy and faster prediction speed, and can meet the requirements of on-line prediction of available bandwidth.
【作者单位】：河南机电高等专科学校计算机科学与技术系;
【基金】：河南省教育厅科学技术研究重点项目(12A520019)
【分类号】：TP393.06

【参考文献】