Design and Implementation of SDN Routing Optimization Algorithm for Electric Power Communication based on Reinforcement Learning

Routing algorithm of SDN controller has an important impact on the performance of SDN. An optimized routing algorithm for SDN based on reinforcement learning is designed in this paper for the communication requirements of low-delay services in electric power communication networks. The main work includes using a reinforcement learning framework to design state, action and reward, achieve optimal routing strategy selection. Simulation experiments using the 14-node NSFNET topology show effectiveness of the proposed algorithm.