The Research Of Map-matching Algorithms In Vehicular Ad Hoc Networks With Low-sample-rate Dataset

The vehicle-related applications are gradually building there importance in the technology renewed world.Vehicular Ad-hoc networks(VANET)as the platform for these applications is paid more attention by researchers.For many VANET applications the localization information is crucial.However,due to various factors,the position obtained by GPS is not accurate,which will have a serious impact on vehicular applications.Map-matching is a technology using digital maps to correct the vehicle location,and provides a new idea for solving VANET positioning distortions.In this thesis,the problem of how to reduce the impact of external environment on the vehicle position is solved with the map-matching technology.To improve the positioning accuracy of positioning system,the main work is as follows:We point out the current research status of map matching give the principle of it.In order to have a clear understanding of the existing map matching algorithm,we classify by its feature.We point out the shortcomings of present algorithms and.In addition,we found the map matching algorithm for high sample rate set will not have a good result for low sample rate set.We proposed,the map matching algorithm for low sample rate set should pay more attention.A map matching algorithm(PSMM)is presented for low sample rate set based on the path chosen.The algorithm mainly analyzes the vehicle trajectory in a real world environment and gives geometry law for vehicles.In addition,the geometric and time characteristics are considered between estimated point and digital road.By using real data for simulation,it shows the positioning accuracy of PSMM is more accurate than other algorithms.A map-matching algorithm based on the weighted topological is proposed,which considers the relevance of the roads for better localization abilities.In addition,an analyze the relationships of localization algorithm estimate points in spatial and temporal space is given in detail.The experiment shows that the proposed algorithm can effectively reduce the positioning error,and can improve the positioning accuracy.The experiment of this paper is programmed by Java to compare with others.