P2P环境下面向不确定数据的kNN查询方法  

kNN Queries Method on Uncertain Data over P2P Networks

作　　者：孙永佼[1] 东韩[2] 袁野[1] 王国仁[1] 

机构地区：[1]东北大学信息科学与工程学院,辽宁沈阳110819 [2]国家海洋信息中心,天津300171

出　　处：《东北大学学报（自然科学版）》2012年第5期632-635,共4页Journal of Northeastern University(Natural Science)

基　　金：国家自然科学基金资助项目(61025007;61073063);国家自然科学基金重点资助项目(60933001);国家海洋公益性行业专项项目(201105033)

摘　　要：由于仪器的不精确和网络延时等原因,在传感器网络和P2P系统中数据都存在不确定性.为解决此问题,基于现有的集中式的不确定数据的kNN查询方法,提出了一种在P2P环境中对不确定数据的kNN查询方法.该方法在super-peer的网络拓扑结构的基础上,以一种扩展的R树(P2PR-tree)作为此查询算法的空间索引结构,解决P2P环境中对多维数据的索引.并且结合两种剪枝策略减小了候选集的范围和减少了查询在P2P网络中的网络代价.实验结果表明,该方法在减少网络代价方面具有较高的性能.There exists an inherent uncertainty on the data objects in sensor networks and P2P systems,due to imprecise measurements and network delays.In order to solve the problem,a novel k nearest neighbor query processing method on uncertain data over P2P networks which is based on k nearest neighbor query processing method on uncertain data in centralized environment was proposed.This method is based on super-peer network topology,and adopts an extended R-tree index,called P2PR-tree,to index the dataset in the distributed database for solving multi-dimensional data index in the P2P environment.Using two pruning algorithms,the number of candidate sets is reduced,and the computation costs and network overhead of kNN queries are further reduced.The experimental results are conducted to verify the high performance of our method on network costs.

关 键 词：SUPER-PEER KNN查询 不确定数据 P2PR-树 全局索引 

分 类 号：TP311.133.1[自动化与计算机技术—计算机软件与理论]