A Combined Network Traffic Prediction Method Based on Ensemble Empirical Mode Decomposition

Abstract

The invention belongs to the technical field of network traffic forecasting, and particularly relates to a combined network traffic forecasting method based on ensemble empirical mode decomposition, including: obtaining original traffic data and performing preprocessing; decomposing network traffic into different The IMF component with a single frequency on the time scale; through autocorrelation and partial autocorrelation analysis, determine the stationarity of the IMF component; predict the smooth IMF component with a linear ARMA model; use the nonlinear Elman neural network for the non-stationary IMF component Network prediction; summing the predicted values ​​of each IMF component to obtain the predicted value of network traffic; the invention describes and predicts actual network traffic more accurately and comprehensively, thereby improving prediction accuracy and increasing prediction reliability.

Description

technical field [0001] The invention belongs to the technical field of network traffic forecasting, in particular to a combined network traffic forecasting method based on ensemble empirical mode decomposition (Empirical Mode Decomposition, EMD). Background technique [0002] In recent years, with the rapid development of the Internet industry, the network scale has become increasingly large and the network structure has become increasingly complex, and network management is facing great challenges. Improving network operation efficiency through an efficient network management mechanism is an important topic for network management workers. [0003] In the face of increasingly complex network interconnection environments and increasing network traffic, researchers and scholars need to use more resources and time to monitor and analyze these network traffic conditions to deal with unexpected situations of network congestion and congestion. to ensure good network quality. Tra...

AI Technical Summary

Problems solved by technology

However, wavelet transform has the problem of determining the number of decomposition layers and choosing wavelet bases, which depends on specific signal characteristics and application fields, and is not self-adaptive.

Images