基于BP神经网络的4G话务预测方法研究

目前常用的话务预测算法主要有移动平均法和指数平滑法，虽然可以达到一定精度的话务趋势预测，但对于短期话务量的周期性变化，例如潮汐效应等场景，传统的话务预测算法存在一定的缺陷。因此需要运用BP神经网络算法，进行更为精准智能的话务预测，以便支撑目前广东公司正在试点的载波动态调度，为潮汐载波调度提供算法依据。本文将对BP神经网络用于话务预测进行研究，并对模型的预测性能进行评价。

Research on 4G Traffic Forecasting Method Based on BP Neural Network

At present, the commonly used traffic prediction algorithms are mainly mobile average method and exponential smoothing method. Although the traffic trend prediction can achieve a certain accuracy, the traditional traffic prediction algorithm has some defects for short-term traffic cyclical changes, such as tidal effects. Therefore, it is necessary to use BP neural network algorithm for more accurate and intelligent traffic prediction, in order to support the current pilot carrier dynamic scheduling of Guangdong company, and provide the algorithm basis for tidal carrier scheduling. This paper will study the application of BP neural network in traffic prediction and evaluate the prediction performance of the model.