Indexing Large Moving Objects from Past to Future with PCFI+-Index Zhao-Hong Liu, Xiao-Li Liu, Jun-Wei Ge, and Hae-Young Bae   View Paper (PDF)   Return to Specialized Indexing Abstract Ideally, moving object databases should handle the past, current and future positions of moving objects efficiently. However, existing indexes such as SEB-Tree, SETI-Tree, 2+3R-Tree, 2- 3RT-Tree and etc. can only provide the capability for past and current query, and the others such as TPR-Tree, and TPR*-Tree can only provide the capability for current and future query. None of them can provide a strategy for indexing the past, current and also the future information of moving objects. In this paper, we present the Past-Current-Future+-Index (PCFI+-Index) which indexes the past, current and future information of the moving objects. The PCFI+-Index builds upon the PCFI-Index which was based on SETI-tree and TPR*-tree. The PCFI+-Index consists of two parts, in memory part with the name frontline, and disk based part. The whole region is partitioned into none-overlapping cells, and a spatial access method is used to index these cells. A set of main-memory TPR*-tree is used to index the moving objects that belong to the cells (one cell, one TPR*-tree). Considering the large update operation triggered by moving objects, the current data file which contains the moving objects' current information is organized as a hash index file. By keeping the restriction in SETI-Index, one page only contains the segments from one cell. Another sparse R*-tree is used to index the lifetimes of the pages. The performance analysis proves that the PCFI+-Index can handle most of the queries efficiently and provides a uniform solution for the trajectory, time-slice, internal and moving queries, and has a better performance than the SETI-Index, TPR*-Index, and PCFI-Index. ©2006 Association for Computing Machinery