摘要： 根据球面上数据对象点的特征和空间数据库查询的需要，给出2种处理球面上最近邻查询的方法，即利用欧氏空间内的空间数据索引结构方法和球面投影于平面方法。在动态密集数据集和动态稀松数据集2种典型情况下分别对该2种方法处理最近邻查询的能力进行分析，结果表明，该2种方法能有效处理球面上具有不同性质特征的空间数据对象点的近邻查询问题。

关键词: 最近邻, R树, TPR*树, TCS树

Abstract: According to the characteristics of the data objects on spherical surface and the needs of spatial database query, this paper proposes two methods for handling Nearest Neighbor(NN) query on spherical surface, which are the query method base on the spatial index structure in Euclidean space named SPINS and the method of spherical surface projecting the plane named CTPS. The performances of the two methods are analyzed with dynamic dense dataset and dynamic sparse dataset. The result shows that the two methods can handle NN query of the data objects with the different properties on spherical surface efficiently.

Key words: Nearest Neighbor(NN), R tree, TPR* tree, TCS tree

中图分类号: 

• TP311.13