基于位置的发布/订阅索引结构  

Location-aware publish/subscribe index structure

作　　者：罗昌银 唐玉茹 李子蹊 李艳红[3] 但唐朋 LUO Changyin;TANG Yuru;LI Zixi;LI Yanhong;DAN Tangpeng(School of Computer Science,Central China Normal University,Wuhan 430079,China;School of Mechanical Engineering,Institut National des Sciences Appliquées de Lyon,Lyon,France;School of Computer Science,South-Central University for Nationalities,Wuhan 430074,China)

机构地区：[1]华中师范大学计算机学院,武汉430079 [2]法国国立应用科学学院机械系,里昂 [3]中南民族大学计算机科学学院,武汉430074

出　　处：《中南民族大学学报（自然科学版）》2019年第2期285-290,共6页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：国家自然科学基金资助项目(61309002);湖北省自然科学基金资助项目(2017CFB135);中央高校基本科研业务费专项资金资助项目(CCNU18QN017,CZZ17003);中南民族大学2017年教研项目(JYX17032)

摘　　要：为了使空间文本数据的处理更加快捷准确,针对基于位置的发布/订阅提出了将R-tree与布尔表达式有效结合的索引结构TR-tree.TR-tree主要由文本索引与空间索引组成,其中文本索引根据订阅中谓词的数量和不同的关键谓词将空间文本数据进行订阅分组和谓词分组.文本索引中使用了操作符列表储存谓词,以达到避免重复储存谓词、减少内存使用的目的.空间索引根据关键谓词与谓词数量构建不同的R-tree,增强了空间修剪的性能.实验结果表明TR-tree具有高效的匹配能力与较好的内存管理性能.In order to make the processing of spatio-textual data faster and more accurate,this paper proposes an index structure TR-tree which combines R-tree and Boolean expression efficiently for location-aware publish/subscribe.The TR-tree mainly consists of text index and spatial index.The text index split the spatio-textual data into subscribe groups and predicate groups according to the number of predicates in the subscription and different key predicate.The text index also uses operator lists to storage predicates,avoiding to repeated storage predicates and reducing memory usage.Spatial index builds different R-trees based on key predicate and the number of predicates,enhancing the performance of spatial pruning.Experimental results show that TR-tree has high efficiency matching ability and powerful memory management performance.

关 键 词：空间关键词 基于位置的发布/订阅 索引结构 布尔表达式 

分 类 号：TP319.56[自动化与计算机技术—计算机软件与理论]