VTPR-tree: An efficient indexing method for moving objects with frequent updates

Moving object databases are required to support queries on a large number of continuous moving objects. Indexes for moving objects must support both query and update operations efficiently. In previous work TPR-tree is the most popular indexing method for the future predicted position, but its frequent updates performance is very poor. In this paper we propose a novel indexing method, called VTPR-tree, for predicted trajectory of moving objects. VTPR-tree takes into account both the velocity and space distribution of moving objects. First the velocity domain is split, and moving objects are classified into different velocity buckets by their velocities, thus objects in one bucket have similar velocities. Then we use an improved TPR-tree structure to index objects in each bucket. VTPR-tree is supplemented by a hash index on IDs of moving objects to support frequent updates. Also an extended bottom-up update algorithm is developed for VTPR-tree, thus having a good dynamic update performance and concurrency. Experimental results show that the update and query performance of VTPR-tree outperforms the TPR*-tree.