Spatial-temporal Indexing Research Based On Road Network

In recent years, the spatial-temporal data management for the urban roadnetwork, has become a spatial-temporal database research areas concern theresearch direction. The spatial-temporal indexing technique be based on theurban road network is the key to handling such spatial-temporal datamanagement problems. How involved massive moving object informationquickly of broad query raised by the client to make the reaction, whichrequires the establishment of efficient spatial-temporal index in the urbanroad network.Spatial-temporal indexing techniques based on the road network, notonly to manage objects’ movements, but also consider the runningenvironment of moving objects, the road network information. Therefore, thedesign of the road network indexing technology and road network model is aclosely related problem, in this study two questions for the content. At thispoint, the background of the urban road network through the existing urbanroad network model is proposed to improve the city road network data model.At the same time, propose a new road network-based spatial-temporal indexFMON-tree. FMON-tree synthesis considers spatial-temporal windowqueries and moving objects global trajectory queries, and the point queryefficiency, it is improved mechanisms for MON-tree. Finally, we throughexperiments comparing the FMON-tree and MON-tree index performance.The main work of this paper can be summarized as follows:1In-depth study of the existing road network data modeling proposeduse of the road network is divided into route+edge space-based strategies,and gives a complete definition associated with the road network data.2In-depth analysis of the classic MON-tree index structure of thecurrent road network, combined with the advantages of the two modelingmethods MON-tree improved FMON-tree, FMON-tree synthesis considerspatial-temporal window queries and global trajectory queries as well aspoint-in-time query efficiency. And, proposed the operationFMON-algorithm.3The use of Network-Based data generating device generates the moving object data set of experimental verification, the experimental resultsshow that FMON-tree in the spatial-temporal window query and globaltrajectory query as well as the point in time query is better than theperformance of MON-tree. However, the performance of index constructedtime need to be strengthened optimization.