基于混合算法的通信用户规模预测方法研究

准确地对通信用户规模进行预测对于通信运营商的决策具有十分重要的意义,而现有的常规预测方法存在预测误差较大、预测速率低等问题。研究一种基于RBF神经网络的通信用户规模预测模型。为了使得RBF神经网络算法预测性能更优,使用梯度下降算法与遗传算法混合对RBF神经网络进行参数优化,提高预测模型收敛效率。实例分析表明,使用本文研究的混合RBF神经网络预测模型的预测结果明显优于其他传统的预测模型。同时,在预测速度上也具有较大的优势。

A communication user scale prediction method based on hybrid algorithm

It is very important for the decision-making of communication operators to accurately predict the scale of communication users. However, the existing conventional prediction methods have problems such as large prediction error, low prediction rate and so on. We study the user scale prediction model based on the RBF neural network, and in order to improve the prediction performance of the RBF neural network algorithm and enhance the convergence efficiency of the prediction model, we combine the gradient descent algorithm and the genetic algorithm to optimize the parameters of the RBF neural network. Example analysis shows that the hybrid RBF neural network prediction model is better than other traditional prediction models, and it has an advantage in predicting speed.