GSPress:一个GML流压缩器  

GSPress:a GML Stream Compressor

作　　者：于一男[1] 关佶红[1] 周水庚[2] 朱付保[2,3] 

机构地区：[1]同济大学计算机科学与技术系,上海201804 [2]复旦大学计算机学院,上海200433 [3]郑州轻工业学院计算机与通信工程学院,河南郑州450002

出　　处：《小型微型计算机系统》2011年第3期397-401,共5页Journal of Chinese Computer Systems

基　　金：国家"八六三"高技术研究发展计划项目(2009AA01Z135)资助;国家自然科学基金项目(60873040)资助;地球探测与信息技术教育部重点实验室开放基金项目(2008DTKF008)资助

摘　　要：GML已成为地理数据存储和交换的标准.在实际应用中,GML数据通常以数据流的形式在网络上传输.由于GML文档含有大量重复标签和高精度浮点数,使得GML文档通常比较大.因此如何有效地压缩GML数据流来减少网络传输的代价就显得越来越重要.本文提出一种GML数据流压缩算法,它只需扫描GML文档一次,将GML文档的结构和数据分离后分别进行压缩.对GML文档中的结构部分采用自适应方法进行压缩;对坐标数据则利用增量压缩的方法压缩.实验结果表明GSPress是一种高效的GML流压缩算法.GML has become the standard for geographical data representation and exchange. In real applications, GML data is always transmitted via network in the form of stream. However, GML documents are usually very large in size because of the verbose tags and large amount of text-encoded coordinate data in the documents. Thus, it is necessary to compress GML stream so as to reduce the transmission cost. In this paper, we propose a GML stream compressor GSPress. It separates spatial data from structural information and compresses them individually. It employs an adaptive method to compress the structural information and the delta compression method to compress the spatial coordinate data. Experimental results show that GSPress can achieve satisfactory compression performance and good tradeoff between compression efficiency and compression effectiveness.

关 键 词：GML 数据流 增量压缩 自适应方法. 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]