Traffic Distribution Forecasting in Packet-Switching Networks

This paper presents a traffic distribution forecasting model in packet-switching networks with mapping these networks into multi-commodity networks. Firstly, the radial basis function (RBF) networks is applied to monitor and learn the real traffic distribution at present time. Then, a quadratic model is used to calibrate these functions for precise traffic distribution forecasting. The implementation of the proposed model is demonstrated through the use of a numerical example.