Multichannel data forwarding decision-making method based on reinforcement learning

Abstract

The invention discloses a multi-channel data forwarding decision-making method based on reinforcement learning, and the method comprises the following steps: S1, system deployment: carrying out the system deployment which comprises a user side access module, an exit routing information collection module and a multi-channel routing management center. The invention to the technical field of computernetworks. According to the multi-channel data forwarding decision-making method based on reinforcement learning, the artificial intelligence model has strong learning ability and good generalizationdue to improvement of equipment computing power and model expression ability, and the perception ability of deep learning and the decision-making ability of reinforcement learning are combined in a universal form; the static routing information is directly controlled in an end-to-end learning manner, the input-to-output direct control is realized in an end-to-end learning manner, the network security and confidentiality are higher, and the network bandwidth is not occupied; when the network topology or link state changes, the static routing information does not need to be adjusted on a large scale, thereby reducing the operation complexity.

Description

technical field [0001] The invention relates to the technical field of computer networks, in particular to a multi-channel data forwarding decision-making method based on reinforcement learning. Background technique [0002] In recent years, with the rapid development of the Internet, many emerging applications, including industrial Internet, remote office, online games and holographic communication, have emerged in large numbers. The emergence of these emerging applications brings highly differentiated service quality requirements on the one hand, and challenges the full utilization of network resources on the other hand. The ever-increasing network requirements of users and the urgent expectation for high-quality service quality have led to the emergence of many technologies, among which configuring multiple egress routes for users is a solution. [0003] In the past, using a single channel for user traffic data transmission encountered network communication link congesti...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies of the prior art, the present invention provides a multi-channel data forwarding decision-making method based on reinforcement learning, which solves the problem that when the network topology or link state changes, the static routing information needs to be adjusted on a large scale, which will greatly increase the operation time. complexity problem

Images