Low-delay high-reliability V2V resource allocation method based on deep reinforcement learning

Abstract

The invention proposes a low-delay high-reliability resource allocation method based on deep reinforcement learning, and NR-V2X side chain resource allocation outside the coverage range of a base station is considered, wherein a vehicle schedules URLLC slice resources used by V2V users in a 5G network according to information observed by the vehicle and a Q network obtained in a training stage. In order to maximize the energy efficiency of V2V communication and ensure the reliability and time delay requirements of communication, a deep reinforcement learning architecture using centralized training and distributed execution is provided, and a model meeting the requirements is trained by means of a DDQN learning method. Modeling of targets and constraints in a resource allocation problem is converted into income design in deep reinforcement learning, the joint optimization problem of V2V user channel allocation and power selection can be effectively solved, and the performance is stable in optimization of a series of continuous action spaces.

Description

technical field [0001] The present invention relates to a vehicle networking technology, in particular to a resource allocation method for the vehicle network, and more specifically, to a low-latency and high-reliability vehicle-to-vehicle (V2V) communication based on deep reinforcement learning resource allocation method. Background technique [0002] Vehicle-to-everything (V2X) is a typical application of the Internet of Things (IoT) in the field of Intelligent Transportation System (ITS). A ubiquitous smart car network formed. The Internet of Vehicles shares and exchanges data according to agreed communication protocols and data interaction standards. It enables intelligent traffic management and services, such as improved road safety, enhanced road condition awareness, and reduced traffic congestion, through real-time perception and collaboration among pedestrians, roadside facilities, vehicles, networks, and the cloud. [0003] Deep reinforcement learning is a kind o...

AI Technical Summary

Problems solved by technology

All the above works do not take into account the energy consumption brought by V2V communication

At the same time, because the resource allocation scheme using the centralized reinforcement learning architecture needs to report the vehicle information to the central controller, the transmission overhead is large, and it increases sharply with the increase of the network size, which makes the method unable to be extended to large networks; In the resource allocation scheme using a fully decentralized reinforcement learning architecture, each agent can only observe part of the information related to itself, which makes the trained model inaccurate

Images