遗传算法优化延迟时间和嵌入维的网络流量预测

为了提高网络流量的预测精度，利用延迟时间（τ）和嵌入维（m）间的联系，提出一种遗传算法优化τ、m的网络流量预测模型（GA-PSR）。将τ和m作为遗传算法的个体，以网络流量预测精度作为目标函数，通过选择、交叉、变异等操作找到最优τ和m值，重构网络流量序列，采用BP网络对网络流量建立单步、多步预测模型。仿真实验结果表明，相对于对比模型，GA-PSR提高了网络流量的预测精度。

Network traffic prediction based on delay time and embedding dimension opti- mized by Genetic Algorithm

In order to improve the prediction accuracy of network traffic, a novel network traffic prediction model（GA-PSR） is proposed based on Genetic Algorithm which optimizes the parameters of Phase Space Reconstruction according to the relation between delay time（τ）and embedding dimension（m）. Delay time（τ）and embedding dimension（m）are coded into individual of Genetic Algorithm while the prediction accuracy of network traffic is taken as the objective function, and then the optimal values of delay time（τ）and embedding dimension（m）are found by the selection, crossover, mutation which are used to reconstruct the network traffic series. BP network is used to establish single, multiple step network traffic pre-diction models by the reconstructed network traffic series. The simulation experiments are carried out on network traffic series, and the results show that GA-PSR has improved the prediction accuracy of network traffic.