嵌入式空间索引策略

谢忠; 凤鸣; 马常杰

嵌入式空间索引策略

中国地质大学信息工程学院，湖北武汉 430074

基金项目:

国家高科技研究发展计划(“863”计划) 2002AA135140

详细信息

作者简介:
谢忠(1968-)，男，教授，博士生导师，主要从事嵌入式地理信息系统和网络地理信息系统的教学与研究工作.E-mail：xiezhong@public.wh.hb.cn

中图分类号: TP311
计量
- 文章访问数: 3496
- HTML全文浏览量: 101
- PDF下载量: 1
- 被引次数: 0
出版历程
- 收稿日期: 2006-05-30
- 刊出日期: 2006-09-25

Index Strategies for Embedded-GIS Spatial Data Management

Faculty of Information Engineering，China University of Geosciences，Wuhan 430074，China

摘要

摘要: 嵌入式GIS的移动终端设备被有限的资源所束缚.使得需要使用合适的策略去解决这些问题如有限存储与大数据量之间的矛盾以及适应实时的需求.分别总结了移动终端和GIS空间数据的特点, 如移动终端的效率性和空间数据的复杂性等.将嵌入式GIS应用划分为3类: 电子地图浏览、导航系统和野外数据采集.综合描述已有的索引算法: R树、四叉树、Buddy树, 同时提出了基于主题导向的空间索引策略.实践证明, 该索引策略对已有的索引算法扬长避短, 大大提高嵌入式GIS数据存取、内存使用和CPU性能方面的效率.
- 嵌入式GIS /
- 空间索引策略 /
- 空间数据管理
Abstract: Mobile terminals with embedded-GIS are constrained by limited resources.Proper strategies are required to solve problems such as limited memory conflicting with large quantities of data and the requirement of real-time processing.This paper firstly analyzes the characteristics of mobile terminals and GIS spatial data, such as the efficiency of mobile terminals and the complexity of spatial data.Then classifies the embedded-GIS applications into three types according to its different themes: electronic-map, navigation system and field data collection.Finally it describes the integration of existing index algorithms (packed R-tree, Quad-tree and Buddy-tree), and advances the application theme-oriented index strategy.The practice proves that this strategy takes advantage of these algorithms and overcomes its shortcoming to improve the efficiency of data access, memory utility and CPU performance.
- embedded-GIS /
- spatial index strategies /
- spatial data management

HTML全文

图 1 网格文件索引

Fig. 1. Grid file index

下载: 全尺寸图片幻灯片

图 2 动态金子塔型的多级索引

Fig. 2. Dynamic pyramidal index

下载: 全尺寸图片幻灯片

参考文献(11)

[1]	Beckmann, N., Kerigel, H., Schneider, R., et al., 1990. TheR*-Tree: An efficient and robust access method forpoints and rectangles. In: Proc. ACMSIGMOD conf., Atlantic City, NJ, USA. 322-331.
[2]	Dreamtech Soft Research Group, 2002. Programming for em-bedded systems. Electronics Industry Press, Beijing. 10-12 (in Chinese).
[3]	Guttman, A., 1984. R-tree: A dynamic index structure forspatial searching. In: Proc. ACMSIGMODconf. on themanagement of data. Waterloo, Canada. 47-57.
[4]	Ning, A., Sivasubramaniam, A., Narayanan, V., et al., 2000. Analyzing energy behavior of spatial access meth-ods for memory-resident data. VLDB Journal, 8: 410-420.
[5]	Raghu, R., Johannes, G., 2004. Database management systems. Tsinghua University Press, Beijing. 719 (in Chi-nese).
[6]	Roussopoulos, N., Leifker, D., 1985. Direct spatial search onpictorial data-bases using packed R-trees. In: Proc. ACMSIGMODconf., Austin, Texas. 17-31.
[7]	Shekhar, S., Sanjay, C., 2004. Spatial databases: A tour. Electronics Industry Press, Beijing. 116-119.
[8]	Zhang, T. Q., Ling, X. K., Yu, X., 2004. The lower powerdesign of SoC system. The Application of SCMand em-bedded system. http://www.51eda.com (in Chinese).
[9]	Dreamtech软件研发组, 2002. 嵌入式系统编程源代码解析. 北京: 电子工业出版社. 10-12.
[10]	Raghu, R., Johannes, G., 2004. 数据库管理系统原理与设计. 北京: 清华大学出版社. 719.
[11]	张天骐, 林孝康, 余翔, 2004. SoC系统的低功耗设计. 单片机及嵌入式系统应用. http://www.51eda.com. https://www.cnki.com.cn/Article/CJFDTOTAL-DPJY200406004.htm