Research On Performance Analysis And Optimization Of 802.11p-based Internet Of Vehicles

Internet of Vehicles(Io V)makes that vehicles can communicate with each other via wireless communication technology to exchange safety message,such as road conditions and kinestates of vehicles and so on.In this case,vehicles can react in time based on the received messages to reduce traffic accidents,ease traffic congestion and save consumption.However,the kinestates of vehicles is time-dependent due to vehicles moving with the high-speed in the complex road environment.In this case,the connectivity among vehicles is dynamic,resulting in the communication performance in terms of packet delay and packet delivery ratio being time-varying.Once the time-varying communication performance of Io V cannot satisfy the basic communication requirement,i.e.,vehicles cannot receive the safety information successfully with required time,road safety would not be guaranteed.Thus,it is necessary to analyze the communication performance of Io V.Vehicle platoon of autonomous driving and vehicular fog computing system are two typical application scenarios of Io V.The number of data generated by vehicles in the Io V has increased dramatically as vehicles are equipped with more and more sensors.Moreover,it is hard for vehicles to process such a huge amount of data in time due to the limit computational capacity of vehicles.To solve the problem,many researchers have proposed to offload tasks in the vehicular fog computing system.It is a challenge to optimize the task offloading scheme while considering the impact of transmission delay,tasks computational requirements,vehicles mobility,and the arrival/departure of tasks.IEEE 802.11 p protocol is widely used in the Io V.This paper focuses on the vehicle platoon of autonomous driving and vehicular fog computing system and analyzes the communication performance and optimizes the task offloading strategy of 802.11p-based Io V.The main contents of this paper are summarized as follows:(1)Analyzing the time-dependent performance of 802.11p-based platooning communications in the presence of disturbance.The key to the safety of vehicle platoon is the string stability.However,the communication performance is time-varying when one vehicle in platoons suffers from a disturbance,affecting the string stability.This section considers the impact of disturbance and constructs models to analyze the time-dependent performance of802.11p-based platooning communications.The connectivity model is first constructed.Then the pointwise stationary fluid-flow approach(PSFFA)is adopted to model the dynamic queuing behavior of the vehicle transmission queue and derive the expression of the average number of packets.Afterwards,the mean and variance of the service time is derived based on the connectivity model to obtain the average number of packets.Finally,the time-dependent packet delay and packet delivery ratio is derived and the numerical results validates that the 802.11 p protocol is able to maintain the string stability under disturbance.(2)Studying the task offloading scheme by considering the multiple computation requirements for the 802.11p-based vehicular fog computing system.Tasks with different computation requirements are transmitted by different access categories queues of 802.11 p,resulting in the different transmission delay.This section optimizes the task offloading scheme for the vehicular fog computing while considering the impact of multiple tasks computation requirements to maximize the system reward.The task offloading problem is first transformed into an semi-Markov decision process(SMDP)model where the system state set,action set,state transition probabilities,and system system reward are defined.Afterwards,the optimal offloading scheme is obtained based on the Bellman equation.Finally,the simulation results show that the proposed offloading scheme performs well compared to the greedy algorithm.(3)Studying the task offloading scheme for the vehicle platoon with 802.11p-based vehicular fog computing system to assist.When tasks are assigned to more vehicles,the computation delay is reduced.However,the transmission delay of 802.11 p protocol is increased.It is a challenge to balance the transmission delay and computation delay of executing tasks to maximize the system reward in the process of tasks offloading.This section considers the impact of 802.11 p transmission delay and optimizes the task offloading scheme for the vehicle platoon with vehicular fog computing system to assist.The task offloading process is first modeled by the SMDP model,and a model is constructed to calculate the 802.11 p transmission delay.Then the value iterative algorithm is adopted to solve the SMDP model to find the optimal task offloading strategy.Finally,the numerical results demonstrate that the performance of the proposed scheme is better than greedy algorithm.