Delay-Sensitive Service For Autonomous Driving Research On Multidimensional Resource Allocation Method For Integrated Communication And Sensing System

This paper focuses on the resource allocation problem of integrated communication and sensing vehicular networks.Given the difficulty of satisfying low latency communication and high accuracy perception in existing networked autonomous vehicle networks,this paper investigates the resource allocation methods for integrated communication and sensing system under different road application scenarios from the perspective of centralized and distributed resource allocation:the centralized resource allocation in urban scenarios with high infrastructure coverage and the distributed self-organized resource allocation methods in areas lacking infrastructure.The contributions of this paper are divided into two main parts.(1)Focusing on the coexistence of delay-sensitive communication services and sensing services in vehicle networks,this paper implements a centralized resource allocation framework for integrated communication and sensing networks based on the Lyapunov stochastic optimization method.The queueing constraint of the delay-sensitive service is transformed and solved by the Lyapunov optimization method,and the communication queue stability constraint is transformed to minimize the upper bound of the Lyapunov drift and penalty function to further realize the joint multi-dimensional resource optimization of the sensing performance and communication performance.Compared with the traditional resource allocation method,the service latency is reduced by 83%.(2)For the integrated communication and sensing vehicular network in areas lacking infrastructure coverage,this paper designs an efficient distributed access algorithm for vehicle users.By establishing a Markov chain model of a multi-priority system,we optimize the access probability of each priority level user under different network loads and number of users.This distributed C&S integrated vehicular network does not need to collect service requests from individual users,which greatly reduces the network control overhead,while the flexibly adjusted user access parameters can achieve high channel utilization efficiency under different network situations.Simulation results show that the proposed multipriority access algorithm can effectively reduce the access delay(about 51%reduction)and improve the channel utilization(about 16%improvement)and sensing mutual information(about 29%improvement)of the integrated communication and sensing system compared to the traditional competing algorithms.