支持频繁更新的移动对象混合索引方法

Hybrid Indexing of Moving Objects with Frequent Updates

(College of Electronic Science and Engineering, National University of Defense Technology, Changsha 410073)

摘要: TPR-tree是目前广泛使用的移动对象当前及未来位置索引技术，但是其频繁更新性能低下.通过在TPR-tree上增加一个指向索引树中间节点的直接访问表(direct-access table)内存结构和建于叶节点之上的Hash辅助索引结构，提出了一种支持频繁更新的移动对象混合索引HTPR-tree，并提出了基于HTPR-tree的扩展自底向上(EBUU)更新算法.性能分析和实验表明，采用EBUU算法的HTPR-tree动态更新性能大大高于TPR\+*-tree等索引，而查询性能仅仅稍逊.
- 移动对象索引 /
- HTPR树 /
- 自底向上更新
Abstract: TPR-tree is the most popular indexing method for the current and future position of moving objects，but its frequent updates performance is very low. HTPR-tree, which is based on TPR-tree, supplemented by a hash index on leaf nodes and a memory-based direct-access table pointing to internal nodes of TPR-tree is presented for moving objects with frequent updates. Also an extended bottom-up update algorithm is developed for HTPR-tree. Performance analysis and experimental results show that the HTPR-tree update performance outperforms any other indexing method including TPR\+*-tree at the cost of slight query performance degrade.
- moving object indexing method /
- HTPR-tree /
- bottom-up update