论文部分内容阅读
网络流量预测是网络管理的重要内容,高效的流量预测方法可提高网络管理效率.针对网络流量的时变性等问题,提出了一种基于智能优化的分布式网络流量预测方法.该方法采用果蝇算法优化3次指数平滑预测模型中的平滑因子,对时间窗口内收集到的网络流量进行预测,从而有效地提高3次指数平滑模型下网络流量预测的准确度与效率.仿真实验表明:相比传统3次指数平滑预测模型,此方法可解决平滑因子的不确定性所导致的预测结果误差问题,有效提高了网络流量预测精度.
Network traffic forecasting is an important part of network management, and efficient traffic forecasting method can improve the efficiency of network management.Aiming at the problem of time-varying network traffic, a distributed network traffic forecasting method based on intelligent optimization is proposed.The method uses fruit fly The algorithm optimizes the smoothness factor in the 3-exponential smoothing prediction model and predicts the network traffic collected in the time window, so as to effectively improve the accuracy and efficiency of network traffic prediction under 3-exponential smoothing model. Simulation results show that: The traditional 3-exponential smoothing prediction model can solve the error of forecasting result caused by the uncertainty of smoothing factor and effectively improve the forecasting accuracy of network traffic.