一种双映射变换的空间索引及空间连接算法研究

A Novel Spatial Index with a High-performance Spatial Join

空间索引会极大地影响空间连接操作的效率。提出了一种基于双映射变换的分布式空间索引,通过结合平面角变换和空间填充曲线的优点,对二维空间进行两次维度变换,使空间数据分片建立在一维的顺序存储队列基础上。在此基础上提出了一种空间拓扑连接算法,并进行了算法的四叉树优化和处理效率实验,对比了本文存储方法和传统R-tree存储在时效性和冗余度方面的效率。实验结果表明,本文方法能支持高效的空间连接。

Spatial indexing seriously impacts the efficiency of spatial processing. In this paper, a new spatial index based on double transformation （DT-index） is proposed. As a dimensional transforma- tion the DT-index benefits from both high and low dimensional mapping. The spatial objects are partitioned in sequential queue;more efficient than area partitioning. A spatial join algorithm based on the DT-index is introduced and optimized with a quad-tree. The experimental results reveal that the proposed method improves the performance of spatial join processing in terms of redundancy and speedup ratio through a comparison with the widely-used R-tree method.