| Vehicle Monitoring System is part of Intelligent Transportation System (ITS), and its application is more and more widly,and users demand the system will provide more features.While the implementation of the features depands on reliable datum and transmitting them in effect,thus it is essential for the system to process datum. Especially because the system is developed based on WebGIS and there are tremendous spatial datum that need to be managed.In the case of that, how to manage spatial data is difficult in the system.Inorder to ensure that the system based on WebGIS can work nomally,so the methods to access data need to be researched.Firstly the general frastructure of the system is introduced,then according to the characteristics of the datum that includes not only attribute data such as vehicles' information and users' information and monitoring datum etc. but also spatial datum such as map information.We designed the databases to save the two kinds of datum respectively so that vehicle monitoring function can be implemented.To improve the ability of parallel access to database and be easier to save datum, the method of save the two kind of data separately,ie the attribute data is saved in relational database in Oracle9i,and the spatial data is saved in Oracle Spatial.And index on map data is created in Oracle Spatial and .tab map information is loaded into database by EasyLoader provided by MapInfo Professinal 8.5.Secondly,Aimed at improve the efficiency of searching spatial data,Squadtree and R tree index arithmetic are mainly researched.After many experiments on performance of building tree,searching data and deleting data,the conclusion is drawn that it is more efficient to use R tree index algrithmetic in the system. So it is adoped and played great role in searching spatial data.At last,the database and the application program were connected by JDBC in Java language.In order to improve the efficiecy of connection and access, database connection pool is applied based on JDBC. |
PDF Full Text Request