摘要
为了解决因业务流迅速增长而造成的网络拥塞现象,提出了一种新的业务流预测算法PSATP(Particle Swarm-ARIMA model based Traffic Prediction).该算法首先利用ARIMA模型和Kalman滤波建立预测方法,并且结合粒子群进行优化,使预测精度得以提高.同时,以实际数据进行仿真实验,深入研究了时延、丢包率与缓冲区、利用率之间的关系.相比于其他算法,仿真结果表明PSATP算法具有较好性能.
In order to mitigate network congestion caused by the rapid growth of traffic, a novel traffic prediction algorithm PSATP （Particle Swarm--ARIMA model based Traffic Prediction） is proposed. In this algorithm, the prediction method is presented by ARIMA model and Kalman filter at first, and the prediction accuracy is improved with particle swarm optimization. Then, a simulation with actual data was conducted to study the relationship between delay, dropping rate buffer and utilization rate. The simulation results show that, compared to other algorithm, PSATP has better performance.
出处
《微电子学与计算机》
CSCD
北大核心
2014年第1期56-59,共4页
Microelectronics & Computer
基金
国家自然科学基金项目(10901144)
关键词
拥塞
预测
精度
粒子群
ARIMA
congestion
prediction
accuracy
particle swarm
ARIMA
