Research On Distributed Resource Allocation And Congestion Control Schemes In NR-V2V

In recent years,mobile communication technology has brought earthshaking changes to daily life,and it is in the rapid development stage of expanding to vertical industries.Among many vertical industries,the Internet of Vehicles industry is considered to be one of the most clearly potential areas.V2X(Vehicle to Everything)technology is a standard technology for vehicle communication and a key technology to ensure the interconnection and intercommunication of vehicle networking.With the gradual advancement of V2X standardization,it has evolved from LTEV2X(Long Term Evolution-V2X)to NR-V2X(New Radio-V2X)version.The constantly evolving V2X technology faces the challenges of the scarcity of dedicated spectrum resources and the ever-increasing performance requirements.How to efficiently use wireless resources has become a key issue in communication research.Based on the NR-V2X standards,this paper focuses on the research of resource allocation mechanism and congestion control mechanism,and combines V2X communication system-level simulation to verify the research content.First,this paper studies the resource selection mechanism in NR-V2V.The vehicle terminal needs to select time-frequency domain resources in the sidelink resource pool to transmit data information.The paper considers the problem of hidden-node caused by the dynamic movement of vehicle terminals and the blocking of buildings,as well as the impact of resource reuse and resource collision.As a result,a two-step distributed resource selection mechanism is proposed.In the system-level simulation,the utility and performance results of the proposed algorithm are analyzed and compared with other schemes to verify the performance advantages of the proposed algorithm.Second,this paper studies the congestion control mechanism in the high-density vehicle scenario.The congestion control problem is modeled by combining the channel busy ratio and the channel occupancy ratio.The basic principle of congestion control is that the vehicle terminal can make effective adjustment strategies immediately after judging congestion to avoid continuous congestion in the resource pool.In this paper,after the simulation analysis of the adjustment of the retransmission times in the 3 GPP standard,a power control algorithm based on deep reinforcement learning is proposed to achieve congestion control.System-level simulation results show that the proposed power control algorithm has obvious performance advantages compared to other schemes.