Multicast Routing Method for Internet of Vehicles in Urban Scenes Based on Reinforcement Learning

Abstract

The invention discloses a multicast routing method for the Internet of Vehicles in an urban scene based on reinforcement learning. The steps are: 1) constructing the Internet of Vehicles; 2) sending HELLO message packets periodically to each vehicle node in the Internet of Vehicles; 3) vehicle nodes Update Q V Q value in the table; 4) Each roadside node in the Internet of Vehicles periodically sends a HELLO message packet; 5) The roadside node updates Q C Q value in the table; 6) The multicast group source vehicle node sends the data packet to the multicast group source roadside node; 7) The multicast group source roadside node forwards the data packet of the destination roadside node to the multicast group member Each roadside node in the roadside node set; 8) The multicast group member roadside node forwards the data packet of the multicast group source roadside node to each vehicle node in the multicast group member vehicle node set. The invention can effectively reduce the end-to-end time delay, and can be used for multicast data routing of the Internet of Vehicles in an urban environment.

Description

technical field [0001] The invention belongs to the technical field of communication, and further relates to a multicast routing method for the Internet of Vehicles in an urban scene based on reinforcement learning in the technical field of network communication. The present invention can be used for multicast data routing of the Internet of Vehicles in an urban scene, and adopts a Q learning method to dynamically select an optimal network node for data transmission. Background technique [0002] The Internet of Vehicles consists of on-board units installed on vehicles and infrastructure units deployed on the roadside, which constitute the basic network unit for Internet of Vehicles communication. Each vehicle node in the Internet of Vehicles can communicate with other vehicle nodes directly or through the existing infrastructure to share information wirelessly. Work collaboratively. Compared with ordinary mobile ad-hoc networks, the rapid movement of vehicles not only lea...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to address the above-mentioned deficiencies in the prior art, and propose a multicast routing method for the Internet of Vehicles in an urban scene based on reinforcement learning, aiming at solving the problem of not making full use of the high bandwidth and stability of roadside nodes in the prior art, resulting in Technical problems that cannot effectively reduce the end-to-end communication delay

It also solves the problem of insufficient adaptability to the high dynamics of the Internet of Vehicles in the existing Q-learning routing method

Images