Deep learning-optical network routing algorithm based on wavelength continuity supervision

In order to reduce the blocking rate of wavelength routing DWDM optical network and improve the wavelength resource utilization, this paper proposes a deep learning optical network routing algorithm based on wavelength continuity supervision (DL-RWA). In this algorithm, wavelength continuity is taken as the key parameter, and the data set is created by supervised learning. After the deep neural network (DNN) is constructed, the data set is used to train it, and the network parameters are adjusted, so that the algorithm can select the routing and wavelength assignment (RWA) scheme with the best wavelength continuity according to the real-time situation of the dynamic network. The simulation results show that compared with the traditional KSP + FF routing algorithm, DL-RWA algorithm can effectively enhance the routing effect and improve the network environment when dealing with the long correlation traffic model (IP traffic simulation).