The Research On Minimizing Distribution Delay Of Cached Data In Vehicular Networks

With the rapid development of vehicular networks,road safety and personal entertainment applications will also increase.A large number of file transfer requests and the uncertainty of vehicle moving location may result in increased transmission delay.Caching popular files on the edge of the network is introduced as an effective method to solve the two problems.Therefore,cache data distribution mechanism in the Internet of vehicles has become a research hotspot in related fields.However,due to high-speed mobile characteristics of the vehicles,the relative speed and distance between vehicles and Road Side Units(RSUs)or between vehicles and other vehicles are constantly changing,which makes it very difficult to maintain a stable wireless communication connection between vehicles and between vehicles and RSU communication,thus affecting the transmission of cache files.In addition,when a cached file can be obtained from multiple source nodes,the most appropriate source nodes need to be selected for transmission to ensure the requirements of low transmission delay and high transmission rate,so that improving the success rate of file transmission.Therefore,this thesis studies the distribution and transmission of cached data in different communication scenarios of vehicular networks,and the main work is as follows.A predictive and hybrid caching transmission scheme for time-delay sensitive services is proposed for a hybrid scenario of vehicle-to-roadside infr-astructure communication and vehicle-to-vehicle communication.the file transmission delay between the requesting vehicle and the RSU or surrounding vehicle is predicted,two communication conditions are modeled respectively,and then predicts the location range that can establish a reliable connection and the location where the connection interruption may occur based on the starting relative distance,the receiving SNR threshold and the relative moving speed.Based on the location information,the delay of obtaining files from RSU and from surrounding vehicles are predicted respectively,including waiting delay and transmission delay.The prediction of waiting delay is obtained by the starting position of the requesting vehicle and the relative speed between requesting vehicle and other nodes.The calculation of transmission delay is based on the average data rate and the effective communication range.The node with the minimum delay will be selected as the data transmission source.Simulation results show that the proposed can significantly reduce the file transfer delay,especially when the requesting vehicle is far from the roadside unit,and the transmission node can be switched from the RSU to the surrounding vehicle to obtain a lower transmission delay.In the context of vehicle-to-vehicle communication,this thesis proposes a Predictive Time Division Transmission(PTDT)algorithm,which aims to reduce the transmission delay of the segmented cached data in the vehicular network and improve the transmission success rate.In order to reduce the calculation time and simplify the prediction process of the delay,the maximum connection time of other vehicles around the requesting vehicle is modeled according to the relative position and speed information.By predicting the data transmission rate of each vehicle at different time,the maximum connection time between each vehicle and the requesting vehicle is divided into multiple slices.On each time slice,the vehicle that can minimize the transmission delay is selected as the transmission source of each file segment.At the same time,the transmission sequence of each transmission source is reasonably arranged,so as to ensure the integrity of files.The simulation results show that the proposed algorithm can reduce the transmission delay of the piecewise cache file effectively and improve the successful transmission ratio of the file.