摘要： 针对GPRS小区流量预测问题，提出一种基于乘积季节自回归移动平均模型的解决方法。该方法分析GPRS小区流量的变化情况，利用小区流量以天为周期变化的特点，应用该模型进行建模，得到GPRS小区的流量变化预测模型。该预测模型可以根据GPRS小区过去的流量情况，预测将来某一时间的流量，为提前采取措施保持GPRS小区性能提供一定的决策依据。
关键词:
GPRS小区,
乘积季节自回归移动平均模型,
流量预测
Abstract: Aiming at the problem of the traffic prediction of GPRS cells, this paper proposes a method based on multiple seasonal Auto- Regressive Integrated Moving Average(ARIMA). The traffic changes of GPRS cells is analyzed in detail, according to the characteristic of cells traffic changing in one day cycle, the multiple seasonal ARIMA model is used to model the process to construct the prediction model of GPRS cells traffic. Based on the previous GPRS cells traffic, the constructed model can predict the traffic and give the decision proof to keep the performance of GPRS cells.
Key words:
GPRS cells,
multiple seasonal Auto-Regressive Integrated Moving Average(ARIMA) model,
traffic prediction
中图分类号:
周鑫, 张锦, 李果, 郑伯峰. 基于乘积季节模型的GPRS小区流量预测[J]. 计算机工程, 2010, 36(18): 76-78.
ZHOU Xin, ZHANG Jin, LI Guan, ZHENG Ba-Feng. Traffic Prediction of GPRS Cells Based on Multiple Seasonal Model[J]. Computer Engineering, 2010, 36(18): 76-78.