无线传感器网络中基于聚类的自适应数据存储  

Clustering-based Distributed and Adaptive Data Storage in Wireless Sensor Networks

作　　者：蔚赵春[1,2] 关佶红[3] 凌鸿[1] 

机构地区：[1]复旦大学管理学院,上海200433 [2]上海浦东发展银行信息科技总部,上海200001 [3]同济大学计算机科学与技术系,上海200092

出　　处：《小型微型计算机系统》2013年第2期304-308,共5页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(60873070)资助;中国博士后科学基金项目(2012M510797)资助

摘　　要：数据存储是无线传感器网络的研究重点.本文分析无线传感器网络中数据存储和访问的相关代价,研究了网状拓扑结构网络中数据存放位置的选择问题.将数据存储问题抽象为传感器节点聚类问题,实现了三种基于聚类的分布式数据存储方法CBDS.为了能够降低能量消耗,CBDS依据生产者和消费者的位置信息、网络拓扑信息及数据速率计算数据存放位置,并且依据这些参数的变化自适应地调整数据存放位置.实验结果表明:CBDS较传统的数据存储方法不仅减少了能量消耗,延长了网络的生命周期,并且降低了访问延迟.Data storage has become an important research issue in wireless sensor networks. In this paper, we analyzed the cost associ- ated with data storage and retrieval, and investigated the problem of data storage nodes selection in mesh topology networks. We for- mulated the data storage problem as nodes clustering. Then we proposed a clustering-based distributed and adaptive data storage strate- gy （CBDS） to decrease the network＇s energy consumption. At the same time, We designed three algorithms to cluster sensor nodes and implemented the distributed data storage strategy. Compared with two existing approaches, our CBDS approach demonstrates sub- stantial performance advantage in extensive experiments. It can not only substantially cut down energy consumption and decrease the delay, but also definitely extends the network＇s lifetime.

关 键 词：无线传感器网络 数据存储 数据速率 聚类 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]