Research On Channel Scheduling Mechanism For IoV Based On RoF-DAS Architecture

With a sharp increase in the number of vehicles,the problems of traffic jam and atmospheric pollution are becoming more serious.Having the ability of improving transportation management efficiency,Internet of Vehicles(Io V)has become a research highlight in the academia and industry field.Io V is confronted with enormous challenges as well as the rapid development.On the one hand,the traffic increasing caused by increasing vehicles desires for higher capacities of storage and processing,on the other hand,the increasing competitions between vehicles would dramatically degrade the network performance of delay and throughput,which will decrease the users' experience.In order to solve the problems,data processing and channel allocation mechanism should be optimized in future Io V.To solve the problems mentioned above,in the second chapter,we studied Radio over fiber-Distributed Antenna System(Ro F-DAS)based Io V architecture,cloud computing technique and fog computing technique at the beginning,respectively.On the basis,we propose a new Io V architecture: Ro F-DAS based hybrid cloud-fog Io V system in the third chapter.This system divides the whole network into three layers,the top layer is responsible for large data storage and processing of delay-tolerant information with cloud computing,and makes dynamic channel scheduling to improve network performance.The middle layer uses fog computing technique to process delay-sensitive data and store a small amount of data.The lowest layer includes Road Side Unit(RSU)and vehicles.Owing to Ro F-DAS,this system can cover a broader range and provide faster transmission rate.Cloud computing effectively increases the speed and capacity of data transmission.Fog computing could support mobility and is closer to users.Therefore,this system increases the ability of processing delay-sensitive traffic.On the basis of the Ro F-DAS based hybrid cloud-fog Io V system proposed in the thesis,a new channel scheduling mechanism which includes initial channel assignment and channel scheduling has also been designed in chapter three.During initialization,the system allocates channels according to the distribution of CCS and RSUs.Related CCS and RSUs are selected according to the requirement of number of users or traffic during the scheduling procedure firstly.Then,we confirm the channel should be add or remove according to the utility function of delay-sensitive information and delay-tolerant information before and after scheduling.On the basis of that,some channels need to switch optimally in order to avoid the influence of great co-channel interference caused by channel reuse after scheduling.Finally,we simulate the proposed network architecture and the channel scheduling mechanism on OPNET simulator in chapter four.Compared with the Road-Based Multi-Interface Multi-Channel(MIMC-Road)assignment method,simulation results demonstrate that the mechanism proposed in this thesis can improve the access delay and reliability for delay-sensitive information,and promote the throughput and fairness for delay-tolerant information at the same time.It is worth noting that in the hotspot area,the enhancement of normalized throughput performance is particularly evident.Hence,for the current problems of Io V,the system and channel scheduling mechanism presented in this thesis has a good application foreground.