Research On High-efficiency Channel Access Technology Based On Real-time Network State In Vehicular Networks

With the rapid development of artificial intelligence,Internet of Things,and highperformance computing technology,the connection requirements of modern automobiles are getting higher and higher when they are becoming more and more intelligent,the concept of vehicle networking has been put forward,and the research and development in related fields have received extremely wide range of attention.Due to the mobility of the vehicle itself,vehicle network communication has the characteristics of complex communication environment,rapid network topology change and frequent access and withdrawal of communication nodes.Because of these features,when more and more complex and complicated data are connected to the vehicle network,it is very important to use efficient communication protocols.In particular,an efficient Media Access Control(MAC)layer channel access control protocol will ensure that the members of the vehicle network access the channel in a timely,efficient and reasonable manner in a highly dynamic topology and in a highly competitive communication environment.Enhancing the efficiency of information transmission is of great significance to improving the safety,reliability,intelligence and traffic management efficiency of modern intelligent transportation.IEEE 802.11 p is a standard protocol designed for low latency and high mobility of vehicular communications.However,its Enhanced Distributed Channel Access(EDCA)mechanism with static Access Category(AC)queue assignment and fixed backoff parameter settings can easily cause emergency messages not to be sent in a timely manner,and non-emergency message to be unresponsive for a long time,which directly leads to problems such as road congestion and traffic accidents,which seriously affects people's daily lives and endangers their life safety.In light of this,we firstly design a dynamic queue assignment algorithm aiming at the defect of the EDCA mechanism.This algorithm pre-estimates the transmission delay of each AC queue and re-allocates the appropriate AC queue for each data frame when access channel to achieve the dynamic partitioning of data priorities.We also design an adaptive backoff algorithm by introducing vehicle state parameters and setting network thresholds for each AC.Then we obtain optimal network threshold combinations under different vehicle densities through extensive experimental trainings.The improved mechanism can dynamically adjust the backoff contention window according to the actual network conditions to ensure that real-time messages can be delivered in time,while the fairness of non-real-time message for channel competition being guaranteed.In this article,we use NS-2 network simulation software to test the performance of the improved EDCA mechanism.Compared with the original EDCA mechanism and an EDCA improvement mechanism A-EDCA(Adaptive EDCA Mechanism),the proposed mechanism has a 48% increase over the original EDCA and a 30% increase over A-EDCA of throughput while it has a 53% decrease over the original EDCA and a 43% decrease over A-EDCA of delay,which shows that the efficiency of vehicular communication has been significantly improved.