Research on metropolitan area network traffic prediction technology based on recurrent neural network
The prediction of Metropolitan Area Network (MAN) traffic is crucial for MAN planning and resource allocation.This study constructs a MAN traffic prediction model based on Recurrent Neural Networks (RNN) to accurately predict MAN traffic.Firstly,by analyzing the factors influencing MAN traffic,the study determines broadband user numbers,network concurrency ratio,average busy-hour traffic of mobile phones,internet usage time of netizens,MAN traffic peak value,and average bandwidth during busy hours as the feature variables of the RNN model.Then,utilizing the memory function of RNN,historical traffic data is used as input to establish a traffic prediction model with infinite memory depth.Experimental results demonstrate that the model can accurately predict the trend of MAN traffic,providing valuable insights for MAN planning and resource allocation.
Metropolitan Area Networktraffic predictionRecurrent Neural Network
万方数据
维普
