蚁群优化神经网络的网络流量混沌预测

摘要

为了网络流量预测准确性,提出一种蚁群算法(ACO)优化BP神经网络(BPNN)的网络流量混沌预测模型(ACO-BPNN).对网络流量时间序列进行重构,将BPNN参数作为蚂蚁的位置向量,通过蚁群信息交流和相互协作找到BPNN最优参数,建立网络流量最优预测模型,并采用实测网络流量数据进行有效性验证.结果表明,ACO-BPNN能够准确刻画网络流量变化特性,提高网络流量的预测准确性.%In order to improve the prediction accuracy of network traffic, this paper proposes a network traffic prediction model based on neural network optimized by ant colony optimization algorithm (ACO-BPNN). The data of network traffic are reconstructed by chaotic theory. The parameters of BPNN are considered the position vector of ants. The optimal parameters are found by ant colony optimization algorithm. The optimal model for network traffic is built and the performance of mode are tested by network traffic data. The simulation results show that ACO-PBNN can describe the change rule of network traffic accurately and can improve prediction accuracy.