基于闭欧拉直方图的空间查询代价模型  

The Cost Model of Spatial Queries Based on Closed_Euler Histogram

作　　者：陈海珠[1] 

机构地区：[1]重庆电子工程职业学院软件学院,重庆401331

出　　处：《软件》2013年第6期61-64,共4页Software

摘　　要：欧拉直方图是空间查询代价估算的一种简便而有效的方法。有许多的研究基于这种方法。但是欧拉直方图对空间对象的统计存在计数错误的问题,以MBR近似描述二维空间对象,文[1]提出了闭欧拉直方图并证明了其统计方法的正确性。文[2]以简单凸多边形近似描述二维空间对象,证明了闭欧拉直方图和欧拉公式同样适用于估算在此描述上的空间选择代价。基于简单多边形的近似描述,改进原有的计数方法,可进一步扩展闭欧拉直方图的使用范围。此外,本文给出了该代价模型的一个应用。Euler histogram is a kind of simple but useful method to estimate spatial queries costs. Much spatial query cost estimations have been done based on such histogram. However, there are some statistical errors in Euler histogram for some distributions of the spatial objects--called the boundary problem. Closed_Euler Histogram is proposed to solve this problem. Most of cost model for spatial queries are presented based on minimum bounding rectangle （MBR for short）, but more accurate approximation is required since the spatial topology relationships based on MBR always do not consist with the ones among the real spatial objects. Based on simple polygon approximation, the spatial query cost can be evaluated more precise through improving the statistical method. Finally, an instance is used to show feasibility of the cost model.

关 键 词：空间查询优化 空间选择 代价模型 欧拉直方图 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]