An Efficient Location Index for the Semantic Search of Moving Objects

In moving object databases, researches on the spatio-temporal access method are very important for the efficient search of moving object location in ITS, LBS, and Telematics. Recently, researches are being made actively on the efficient management of the current location of moving objects and on the estimation of future location using information such as the current location and moving pattern of moving objects. In this paper, we propose Map-Based Rtree(MBR-tree), which is a new current location index structure for indexing the current location of moving objects in an urban area, a 2-dimentional space. MBR-tree is an index which forms the MBR(Minimum Bounding Rectangle) of R-tree nodes using static objects(or fixed objects) on the map. Because moving objects generally moves within a static object, if the MBR is formed using static objects, we can reduce the cost of updating the index of the current location of moving objects. In addition, it shows superior performance in semantic search that searches in a specific building or place (e.g. “Who are in Konkuk university?”) rather than in an arbitrary area. Finally, to test the index proposed in this paper, we compared its performance with that of hashing technique and Lazy Update R-tree using various datasets and proved the superiority of its performance.