Research On Blockchain-based Resource Allocation Of UAV-assisted Internet Of Vehicles

With the rapid growth of the number of vehicles connected to the Internet of Vehicles,the traffic volume of communication has surged,and the problem of scarcity of spectrum resources has become more and more prominent,posing new challenges to the reliability and real-time performance of vehicle network communication.UAV-assisted Io V takes advantage of the high mobility and ease of deployment of UAVs to offload business flows,reduce interference from ground communication systems,and solve the problem of inability to provide wireless connectivity to users outside the coverage area of base stations.Its deployment improves the quality and performance of in-vehicle applications,but there is still a shortage of centralized management in this networking method.In order to solve this problem,this thesis proposes a blockchain-based UAV-assisted Io V resource allocation method.In the case of few available channels,the spectrum is fully reused to build a UAV-assisted Io V block.The chain resource management model is used to realize the spectrum allocation and power control of the UAV.Blockchain is a new type of decentralized technology.Using blockchain to manage wireless spectrum resources can effectively improve the allocation efficiency of spectrum resources and system reliability,and reduce system maintenance costs.In this framework,how to achieve consensus among nodes while guaranteeing the performance of the vehicle system is a major challenge.The simulation results show that the UAV-assisted Io V resource allocation scheme based on blockchain can improve the reliability of the system,and has effectiveness and advantages in improving resource utilization and reducing delay.The main work of this thesis includes the following points:1.Introduce the contradictory problem of the surge in the number of connected vehicles and the shortage of spectrum resources,and propose solutions to improve the performance of spectrum resource management,mainly including channel allocation and power control of spectrum resources;2.Build the model of the UAV-assisted Io V system,and derive the transmission rate,outage probability and automatic retransmission delay expressions;3.The Ripple consensus mechanism is used for consensus voting on spectrum and power,data is backed up at each node,and UAV nodes jointly maintain the blockchain,which improves the security and reliability of the system;4.Research the resource allocation with the optimization goal of maximizing the average energy efficiency while satisfying the constraints of the network.Use the single-agent Dueling DQN algorithm to solve the optimization problem,and then broadcast the solution results to the blockchain nodes for consensus voting.Compared with the scheme without blockchain,the scheme proposed in this thesis achieves higher average energy efficiency.At the same time,the algorithm adopted in this scheme is superior to the Q-learning algorithm and the random allocation method in performance;5.Research the spectrum allocation and power control with the optimization goal of minimizing the average delay.The single-agent solution reinforcement learning problem is extended to a multi-agent Dueling DQN algorithm to solve the optimization problem,and the voting results of the blockchain are fed back to the reward.Apply the dynamic game of Stackelberg equilibrium to realize the optimal strategy of the whole system.Simulation experiments show that the multi-agent reinforcement learning scheme is better than the single-agent scheme in terms of average delay,and the proposed scheme is effective and superior.