SDN route planning method based on reinforcement learning

Abstract

The invention discloses an SDN route planning method based on reinforcement learning. The method comprises the following steps: constructing a reinforcement learning model capable of generating a route by using Q learning in reinforcement learning on an SDN control plane, designing a reward function in the Q learning algorithm, generating different reward values according to different QoS levels of traffic; inputting the current network topology matrix, traffic characteristics, and QoS levels of traffic in the reinforcement learning model for training to implement traffic-differentiated SDN route planning, and finding the shortest forwarding path meeting the QoS requirements for each traffic. The SDN route planning method, by utilizing the characteristics of continuous learning and environment interaction and adjustment strategy, is high in link utilization rate and can effectively reduce the network congestion compared with the Dijkstra algorithm commonly used in traditional route planning.

Description

technical field [0001] The invention relates to the field of network communication technology and reinforcement learning, in particular to an SDN route planning method based on reinforcement learning. Background technique [0002] The continuous growth of Internet traffic data has caused problems such as a sharp increase in bandwidth consumption, difficulty in ensuring service quality, and increased security issues. The Internet is inseparable from all walks of life and has obviously become the industry with the most promising prospects. With the growth of business, all walks of life and individual users will generate tens of thousands of network information flows every day, such as file transfer, voice call, online game, etc. New application models and requirements are constantly emerging, and the traditional network architecture has been unable to cope with the rapid The developing Internet is facing many problems such as insufficient network address space, increasingly bl...

AI Technical Summary

Problems solved by technology

However, if the forwarding of all data packets only depends on the shortest path algorithm, it will cause a serious problem. Data flows are easily gathered together because of choosing the same forwarding path, which greatly reduces link utilization and also It is easy to cause network congestion

Some multi-path protocols that exist also do not consider the quality of service (QoS) requirements of different traffic flows. From the perspective of path optimization, this is limiting because it does not consider the traffic status of the entire network.

Images