An Efficient Routing and Spectrum Assignment Algorithm Using Prediction for Elastic Optical Networks

Holding time awareness provides the additional information that could be used in routing and spectrum assignment (RSA) in elastic optical networks, and thus improves the performance of RSA. We propose an RSA algorithm that achieves low blocking probability by using Back Propagation neural network (BPNN) to predict the information of each future request with holding time awareness. We examine the performance of the proposed algorithm through simulation experiments, and the results show that the proposed RSA algorithm obtains better performance compared to existing RSA algorithms without the usage of prediction.