A NN-GM (1,1) model - based analysis of network traffic forecasting

For the method of neural network can modify and better the forecasting effects of grey prediction model, a combined NN-GM (1,1) forecasting model is proposed. After the time sequence of the network traffic was analyzed, the equidistant metabolic GM (1,1) forecasting model was constructed, and the residual error of the model was corrected by neural network based on Error Back Propagation (BP) algorithm. The forecasting results of the experiments and the simulation show that the combined model is more effective than the single common grey model. The application of the proposed combined model to network traffic forecasting may offer scientific rationale for LAN virus detection, illegal invasion prevention and Internet routing decision.