摘要
在R树插入算法中采用全新的节点选择算法,一改传统的从根节点开始自上而下的节点选择方案,而是从叶节点层开始,先自下而上再自上而下地选择叶节点,较好地解决了同层节点重叠所导致的查询效率低下的问题。实验证明,提出的R树空间索引方法,不仅在查询效率上明显优于R*树,而且R树生成的时间开销也减少了50%左右,综合性能超过了R*树,便于扩展到三维甚至多维空间中,以实现对空间数据和时空数据的高效查询功能。
Spatial index is a key technique in the field of spatialdatabase. This paper presented a brand-new node-choosing algorithm in the insertion procedure of R-tree spatial index, which was greatly different front the classical algorithm. From the beginning of the leaf node layer, firstly from bottom to top then inversely, chose the right leaf node, and this scheme could solve the problems caused by node overlapping. A comparative performance analysis on the current and improved methods show that not only the query performance but also the generation performance of the improved method is higher than the current method such as R -tree. Moreo＇~er, the impro＇ced method can be extended into the field of muhi-dimension applications such as spatio-temporal application.
出处
《计算机应用研究》
CSCD
北大核心
2008年第10期2946-2948,2955,共4页
Application Research of Computers
基金
国家自然科学基金资助项目(40571121)
江西省教育厅科技计划资助项目(GJJ08152)
香港"数字志莲净苑木构佛寺演示系统"项目
