基于人类周期活动的网络流量线性预测模型
network data flow prediction liner model based on human peridical behavior
在这篇文章中，提出了一个基于人类周期性活动的网络流量预测模型，在因特网的不断发展中，对网络的要求越来越高，除了传统的提升网络的传输速度以为，智能网络的概念也不断提出，智能网络要求对网络的流量未来趋势有一定的预测，对此提出了许多流量预测模型，与其他模型不同，该线性模型基于网络的弱周期性质，短时间连续，这些性质是由于人类的周期活动造成的，因此该模型包含了两方面，在本文中采集了一个典型LAN网络，用户数量大于在100左右的关键点的几个月进出流量数据，与传统的ARIMA模型进行精度比较，体现本论文提出的线性模型拥有更高的精度，而且计算简单方便。
In this paper, advanced a network dataflow liner model based on human periodic behavior.With the develop of Internet,more and more demand appear，beside the tradition method that increasing transmission rate.Smart network appear,smart network want to know furture dataflow date.Many prediction model is proposed for Network dataflow. Different with other model,this model based on dataflow character that weakly periodic, consecutive in an interval casued by human periodic behvaior. Thus the model contain two parts reflect above character. We collect a LAN of 100 users input and output data rate in several months. With this data we compare with tradition ARIMA prediction model to show this model can offer substantial improving in precise and easy caculate.
刘武明、纪越峰、陆月明
通信
通信网络技术网络流量预测人类周期活动
ommunication Network TechnologyNetwork dataflow predictionHumanity periodic behavior
刘武明,纪越峰,陆月明.基于人类周期活动的网络流量线性预测模型[EB/OL].(2011-09-28)[2025-08-02].http://www.paper.edu.cn/releasepaper/content/201109-347.点此复制
评论