Research On Adaptive Moving Object Indexing Technology For Urban Road Network

With the rapid development and application of computer and wireless communication technology,the popularity of wireless mobile terminals and global positioning system,making more and more services based on moving object location to become a reality.The moving object database is the base for effectively supporting the location-based services,which stores the location information of the moving object and processes the spatio-temporal data that changes frequently.In the moving object database,the moving object indexing technology is the core of data management to support the moving object database,which can effectively support the query of moving objects.By studying the existing moving object indexing technology,this paper proposes a method to manage a large number of moving objects.The core of this method is a Hot-spots Adaptive Index(HAI).HAI is a double-level index structure.The upper layer adopts R*-tree to manage the road networks,and the lower layer uses the R-tree group to index the moving object movement information updated in real time.HAI uses a memory-based index structure to manage hot-spots and moving objects in the area,and the non-hot-spot area and its moving objects are managed by an index structure based on the external memory.HAI can carry out the migration of the index according to the change of the vehicle density on the road,so as to guarantee the index update and query performance under the limited memory condition.At the same time using the buffer to save the moving objects information to index the bulk update,reducing the moving object frequently updated I/O operations,with good scalability.Through theoretical analysis and experiments,it is shown that HAI not only reduces the I/O cost of index creation and maintenance,and also greatly improves the efficiency of querying moving object trajectories compared to NDTR-tree(Network-constraint Moving Objects Dynamic Trajectory R-tree).