基于网格分组移动对象的索引方法  

An Indexing Method for Grouping Moving Objects with Grid

作　　者：梁野[1] 龚雪晶[1] 慈林林[1] 

机构地区：[1]北京理工大学计算机科学技术学院,北京100081

出　　处：《北京理工大学学报》2008年第8期692-696,共5页Transactions of Beijing Institute of Technology

基　　金：国家部委预研项目(41316040104)

摘　　要：针对目前使用广泛的移动对象索引技术对受限范围内的海量数据索引管理问题.引入了网格技术实现移动对象的分组管理,提出了基于网格分组的移动对象GG TPR树索引方法.该方法利用网格分组的方式,将位置与运动行为接近的移动对象进行整体维护和批量管理,从而减少索引维护的中间环节,提高移动对象的索引效率.模拟实验结果表明,利用GG TPR树索引受限范围内的海量数据较已有的移动对象索引方法性能更优.TPR-tree is the most popular indexing method for moving objects, but it is inefficient to deal with a huge number of moving objects within a limited area. When taking cognizance of the regular track of moving objects within a limited area, the grid technique can be relied upon to group the moving objects. With the GG TPR-tree, grouped moving objects which are neighbors and will run to the same direction in the future to improve the efficiency in indexing. Experimental results showed that GG TPR-tree＇s performance was better than the other indexing structure on managing a great number of moving objects within a limited area.

关 键 词：移动对象 网格分组 TPR树 索引算法 

分 类 号：TP393.103[自动化与计算机技术—计算机应用技术]