Research On Hybrid V2V And V2I Transmission Patterns In Vehicular Networks

Nowadays,vehicular networks(VNs)have been undergoing an unprecedented development.With the explosive growth of the amount of data,the reasonable data transmission patterns and effective evaluation on data transmission performance have became research hotspots.In VNs,Dedicated Short-Range Communications(DSRC),as a wireless communication technology,provides two patterns,i.e.,Vehicle-to-Vehicle(V2V)communication and Vehicle-to-Infrastructure(V2I)communication to support data transmission.V2V and V2I are complementary while promoting data transmission efficiency collaboratively in VNs.Transmission time and handover cost are the two key performance metrics to evaluate the data transmission performance.In practice,the movement of vehicles has both randomness and regularity in multilane scenario.Vehicles will have frequent infrmation exchange and link handover due to different speeds.Based on the same-direction and reserve-direction multilane scenario together with different data service types,this thesis analyzes the exact result on average transmission time and the average number of handovers under V2I and V2V patterns.The main contributions of this thesis are summarized as follows:In the same-direction multilane scenario,the key parameters that influence hybrid V2V and V2I average transmission time are explored.The average transmission time,average download time and average waiting time under hybrid V2V and V2I patterns are derived based on probability theory and classification methods.The challenge in analysis is the correlation between V2V and V2I transmission process and the randomness of the vehicle’s location.Simulation results demonstrate that vehicle density,RSU uncoverage range and cached probability are the key parameters which influence transmission time.Meanwhile,the accuracy of theoretical analysis is verified.In the same-direction multilane scenario,aming to solve the frequent handover problem caused by spatial position variance of vehicles in V2V transmission process,the Maximum Single Download Time(MSDT)handover strategy is proposed to realize the minimum number of handovers in V2V transmission process.Theoretical analysis on its performance is made based on conditional probability theory and convolution method.The difficulty is the correlation between several successive handover process.Simulation results demonstrate that MSDT handover strategy reduces the number of handover significantly in the same time cost compared with random selection strategy under unit disk model(UDM)and log-normal fading model.Meanwhile,the accuracy of theoretical analysis is verified.In the reverse-direction multilane scenario,this thesis focuses on the choice on V2V or V2I patterns when the message sends back,V2V asssited strategy and V2I wait strategy are proposed.Theoretical analysis on the time cost of the above two strategies is made based on the spatial distribution feature of vehicles and probability theory.The key parameters that influence the failure ratio of V2V assisted strategy and initial position of the assist vehicle are also explored.Simulation results compare the advantages and disadvantages of V2V asssited strategy and V2I wait strategy in terms of time loss respectively.Vehicle density and vehicle speed are demonstrated as influential factors of the failure ratio of V2V assisted strategy and initial position of the assisted vehicle.Meanwhile,the accuracy of theoretical analysis is verified.