摘要
本文提出一种融合小波变换和最小二乘支持向量机(WT-LSSVM)的组合话务量预测模型。对呈现非平稳特性的话务量数据先用Mallat算法进行分解,得到低频分量和高频分量,然后对低频分量和高频分量进行单支重构,再对重构后的各分量分别用LS-SVM模型进行预测,最后合成话务量,实验表明该组合预测模型有较高的预测精度和稳定性。
This paper presents a traffic prediction model that combines wavelet transformation and leaset squares support vector machine（ WT-LSSVM）. Firstly, decompose non- stationary traffic data with Mallat algorithm and get low frequency component and high frequency component. Secondly, reconfigure each component and use IS - SVM model to predict each reconfigured one. Then the traffic can be achieved. The results of experiments have testif - ied higher prediction accuracy and stability of this combined traffic prediction model.
出处
《激光杂志》
CAS
CSCD
北大核心
2013年第4期40-42,共3页
Laser Journal
基金
中国移动通信集团新疆有限公司研究发展基金项目(项目编号:XJM2012-01)
