The Software-Defined Optical Network(SDON)is the latest trend and orientation of the development of optical networks.In order to further improve its intelligentization and dynamic support to traffics
it becomes increasingly important to predict the longer-term traffic flow on the basis of the short-term traffic flow behavior.This article proposes a Bayesian Modelbased Traffic Prediction(BM-TP)scheme
which makes full use of the model to make long-term prediction of SDON traffic flow.Simulation results verify the accuracy and effectiveness of this scheme.