基于向量自回归模型的移动通信基站流量预测

发布时间：2018-03-10 00:11

本文选题：城市基站　切入点：流量预测　出处：《工业工程与管理》2017年04期 　论文类型：期刊论文

【摘要】：城市移动通信基站流量的准确预测对于关键基站的拥堵控制、基站新址的选择有着重要作用。基站流量数据不仅是区域的静态表现,同时也反映区域人员的流动特性。基站流量具有非线性混沌特性,而传统的线性时间序列方法比如自回归移动平均模型难以有效地捕获实际基站流量序列中复杂的非线性因素。同时,仅考虑单个基站时间序列而忽略邻近基站的影响并不能反映基站流量的动态特征。基于向量自回归模型(VAR)对大规模基站流量数据进行整体分析,将多响应变量预测问题转化为单响应变量预测模型,运用Lasso变量选择方法筛选目标基站的重要关联基站。实例表明,相对于传统预测方法,VAR-Lasso类方法不仅提高了基站流量的预测精度,同时也实现了大规模基站的实时预测。
[Abstract]:The accurate prediction of base station traffic in urban mobile communication plays an important role in the congestion control of key base stations and the selection of new site of base station. The traffic data of base station is not only the static performance of the region. At the same time, it also reflects the flow characteristics of people in the region. However, the traditional linear time series methods such as the autoregressive moving average model can not effectively capture the complex nonlinear factors in the actual base station traffic sequence. Considering the time series of a single base station and neglecting the influence of adjacent base station, it can not reflect the dynamic characteristics of base station traffic. Based on the VAR-based vector autoregressive model, the traffic data of large scale base station are analyzed as a whole. The multiple response variable prediction problem is transformed into a single response variable prediction model, and the Lasso variable selection method is used to screen the important associated base stations of the target base stations. Compared with the traditional prediction method, the VAR-Lasso method not only improves the accuracy of base station traffic prediction, but also realizes the real-time prediction of large scale base station.
【作者单位】：上海交通大学机械与动力工程学院;
【基金】：国家自然科学基金面上项目(71672109)
【分类号】：TN929.5

【相似文献】