Research On Traffic Data Perception Algorithm Based On Edge Computing Vehicle Network

With the rapid development of urban traffic network system,traffic data is increasing day by day.In recent years,the edge computing architecture places computing,storage and network resources on the edge nodes of the network,which greatly reduces the delay of data processing,and provides a effective solution for real-time traffic data processing.In addition,compressed sensing has become a popular technology for reducing data transmission,which can be used to reduce the amount of data transmission and reconstruct the original data,so as to obtain the relevant statistical information of the road,so as to better monitor the urban road and reduce the traffic pressure.The development of these technologies plays a key role in the related processing of traffic data.Generally speaking,the existing problems are as follows: the existing data processing based on cloud computing architecture is too late to meet the real-time data processing;with the rapid development of the traffic network,the amount of data generated will grow exponentially,and the bandwidth resources will become the bottleneck of data upload;the current data processing architecture and algorithm,the processing of big data sets will exist for a long time Recovery accuracy and other issues.This paper takes traffic state estimation as the research object,and the main work is as follows:1.The traffic sensing system in mobile edge computing vehicle network is studied.Each mobile edge computing server can collect and recover traffic data in its local server.On this basis,this paper proposes a bandwidth constrained traffic sensing problem to minimize the estimation error based on the collected traffic data.In order to solve the problem,this paper first proposes a bandwidth aware data collection algorithm,which selects the best upload traffic data by evaluating the priority of each road section covered by the mobile edge computing server.Then,this paper proposes a data recovery algorithm based on convex optimization,which minimizes the estimation error by converting the bandwidth constrained traffic perception problem to the norm minimization problem.Finally,the simulation model is implemented and the performance is evaluated.The simulation results verify the superiority of the algorithm.2.The data sensing system based on stack autoencoding model in mobile edge computing vehicle network is studied.Each mobile edge computing server is responsible for data collection and processing,while the cloud layer is responsible for data recovery.In order to verify the feasibility of stack autoencoding model in the data perception and recovery of the Internet of vehicles,the data deviation,sensing range and data sampling rate are changed,and the convex optimization algorithm using interior point method and alternating direction multiplier method are compared.The simulation results verify the superiority of stack autoencoding model in data recovery algorithm.The related work of this paper provides a feasible research idea and technical route for the research and application of the data perception of the Internet of vehicles,and has a certain reference and application value for the follow-up in-depth study of related directions.