Research Of Intelligent Transportation Architecture Based On SDN

With the acceleration of urbanization,the scale of traffic management is constantly expanding,and the number of motor vehicles is also increasing day by day,which makes the demand for roads increase sharply.But most of the existing traffic management system is decentralized management traffic,not formed a complete system architecture,department information can’t reasonable use and not for the transportation scheduling,and introduces advanced technology in intelligent transportation,and with the development of wireless network and artificial intelligence technology and technology has been used in the fields of computer technology in intelligent transportation.In view of this,it is considered to apply the concepts of SDN(Software Defined Network),such as transfer control separation and centralized unified control,to the intelligent traffic management system.SDN is a new type of network innovation architecture,which has the characteristics of rotating control separation,centralized control,open programming and so on.It can manage the network uniformly,realize multi-layer collaboration,centralize control resources,and simplify the underlying applications.The paper compares the similarities between software-defined network and intelligent transportation system,and draws on the characteristics of SDN,the intelligent transportation architecture system based on SDN is designed.At the same time,this article combines path guidance,dynamic speed limit,intersection signal light coordination,etc.It applied to practical application scenarios of adjacent intersections,aiming to improve traffic management,increase road use efficiency and increase the transit time of cars.Firstly,the paper analyzes the problems in the current traffic management system,such as decentralized management and the inability of multi-department information to be used effectively in real time.Secondly,SDN architecture is introduced,and combined with its advantages and characteristics,an intelligent transportation architecture based on SDN is designed according to the similarity between SDN architecture and intelligent transportation architecture,and the architecture system is discussed in layers and applied to adjacent intersections.The influence of signal light timing,path guidance and dynamic speed limit on the queuing length at the intersection and the delay time during vehicle driving is analyzed in detail.Under the condition of guaranteeing the passage of each road section and variable speed control,the surrounding traffic flow is controlled according to the shortest travel time to minimize the impact cost.Finally,this paper chooses SUMO to build a road network to simulate road traffic and OMNET++ to do network simulation,which realizes two-way coupling of simulation software.The influence of the improved traffic model on the current traffic system and traffic performance is analyzed through simulation experiments.The results show that the proposed architecture can improve the status of decentralized management of traffic management,enhance road utilization rate and smooth traffic flow.The simulation experiment shows that compared with the traditional traffic architecture,the collaborative control under the proposed architecture has shown effective improvement in multiple indicators,smoothed the traffic flow,reduced the average delay at the intersection to20.7%,and reduced the average queue length by 35%.