Data Dissemination And Performance Analysis In Vehicular Networks

At present,the world is experiencing the 4th industrial revolution.With the rapid development and deep integration of the new generation of information technologies,such as Artificial Intelligence(AI),big data and the 5th Generation Mobile Networks(5G),autonomous driving has become a global innovation hotspot.The technical route of single-vehicle intelligence,represented by Tesla,mainly relies on the perception and decision-making function of the vehicle itself,and cannot solve the driving safety problems under complex and changeable circumstances.Thus,it needs not only smart vehicles,but also intelligent roads,real-time clouds,reliable networks and accurate maps.The technology route of connected-vehicles intelligence emphasizes the collaborative interconnection of vehicles,roads,cloud platforms and other transportation participants,and vehicular networking is a significant enabling technology to achieve this goal.In a dynamic traffic scenario,how to ensure the effectiveness and reliability of information communication is still facing great challenges.In addition,the world is currently in a critical period of vehicular network deployment,and the guidance theories for network deployment are urgently needed.High-speed mobility and variable network topologies are the main features of vehicular networks,and the movement of vehicles in practice is affected by road layout and traffic flow state.However,for the same reason,compared to the random mobility of nodes in other mobile networks,the movement of vehicles often exhibits some regularity,especially in a sparse scenario.By fully excavating the characteristics of vehicle movement and space distribution,it can provide effective theoretical support for data dissemination and performance analysis in vehicular networks.Based on different application scenarios and network architectures,and focusing on Quality of Service(Qo S),delivery delay and coverage probability,the dissertation conducts a research on data dissemination and performance analysis in vehicular networks by means of model construction,algorithm design,theoretical analysis and performance evaluation.The contributions of the dissertation mainly include:1)For the multimedia data in highway scenario,in order to describe the Qo S of video for vehicle users,the utility function is adopted to map the data volume to the user satisfaction level.To guarantee the basic Qo S,video data requirements are divided into inflexible data requirements and flexible data requirements.With the constraint of limited channel resources,considering the sojourn time of vehicles in the coverage of Roadside Units(RSUs)and multimedia data requirements,the dissertation designs a real-time Access Control and Scheduling(ACS)algorithm for video data.The simulation results show that the proposed algorithm can improve the admission probability and total utility with low computation complexity.2)For the sudden safety messages of traffic accident in highway scenario,considering the abnormal traffic flow in the lane where the accident is located and using the auxiliary communication function of traffic flow in the opposite lane,the dissertation establishes a mathematical model of vehicle mobility and proposes two mechanisms for message forwarding,namely normal “store-carry-forward” mechanism and decelerating “store-carry-forward” mechanism.According to the clustering characteristic of stable traffic flow,the dissertation analyzes the delivery delays of the two mechanisms,respectively.Through simulation experiments,the theoretical results are verified.In addition,the impacts of several relevant parameters on average delivery delay are studied,and the analysis results have reference value for guiding the deployment spacing of RSUs.3)For the heterogeneous Cellular Vehicle-to-Everything(C-V2X)network in urban scenario,considering the coexistence of cellular link and direct link,users have priority to access nearby vehicles in order to provide timely feedback on their safety messages.The dissertation analyzes the coverage probability of four links(i.e.,downlink and sidelink of planar user,and downlink and sidelink of linear user)under vehicleprioritized association scheme.The conditional coverage probability of four links is also derived to study the reliability.The theoretical results are verified by simulation,and the impacts of key parameters on the coverage probability are also studied.The analysis results can provide theoretical guidance for network deployment.4)For Vehicle-to-Vehicle(V2V)communication of C-V2 X network in urban scenario,cooperative communication with cellular relay can avoid link interruptions caused by the large distance between vehicles.Based on two decoding schemes(i.e.,Selection Combining(SC)transmission and Maximum Ratio Combing(MRC)transmission),the dissertation derives the expressions for joint success probability in two consecutive phases considering the correlation of road topology and nodes’ locations.The theoretical results are verified by simulations,and the impacts of several key parameters on the success probability are also studied.The analysis results can provide useful guidance for relay selection and network design.