计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (21): 121-125.

• 网络、通信与安全 • 上一篇    下一篇

无线传感器网络中基于预测的时域数据融合技术

回春立1,2,崔 莉1   

  1. 1.中国科学院 计算技术研究所,北京 100080
    2.中国科学院 研究生院,北京 100080
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-07-21 发布日期:2007-07-21
  • 通讯作者: 回春立

Forecast-based temporal data aggregation in wireless sensor networks

HUI Chun-li1,2,CUI Li1   

  1. 1.Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100080,China
    2.Graduate University of the Chinese Academy of Sciences,Beijing 100080,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-07-21 Published:2007-07-21
  • Contact: HUI Chun-li

摘要:

数据融合是无线传感器网络中重要的研究领域之一。在无线传感器网络中,数据融合的作用主要体现在节省能量、提高数据收集效率、增强数据准确性以及获取综合性信息等几个方面。时间序列分析是一种根据动态数据揭示系统动态结构以及规律的统计方法。基于监测数据的时间序列模型以及时间序列预测,提出适用于无线传感器网络的基于预测的时域数据融合方法,以部署于故宫博物院的环境监测网络采集的温度数据作为样本,通过仿真对该方法进行有效性验证以及性能分析。结果表明,一阶自回归预测算法与其它预测算法相比,具有更好的适用性,当误差阈值为0.05 ℃-0.50 ℃时,预测成功率为21%-83%;当误差阈值为0.05 ℃时,节能收益达到68%。

关键词: 无线传感器网络, 预测, 数据融合, 能量

Abstract: Data aggregation is an important research area in Wireless Sensor Networks(WSN).In WSN,using data aggregation technique can bring the following benefits:saving energy,improving data gathering efficiency,enhancing data accuracy,getting integrated information and so on.Time series analysis is a statistical method which is used to reveal dynamic architecture and changing rule of certain system according to dynamic data.In this paper,a forecast-based temporal data aggregation technique is proposed based on time series model and time series forecast method.The effect and performance of the method is verified and evaluated by simulation using the temperature data collected by an environment monitoring network deployed in Forbidden City.The results show that autoregressive (AR) algorithm is more effective than others for WSN.When error threshold is 0.05 ℃ to 0.50 ℃,forecast success ratio is 21% to 83%.When error threshold is 0.05 ℃,energy saving is up to 68%.

Key words: wireless sensor networks, forecast, data aggregation, energy