XJIPC OpenIR  > 多语种信息技术研究室
Thesis Advisor李晓
Degree Grantor中国科学院研究生院
Place of Conferral北京
Degree Discipline计算机应用技术
Keyword无线传感器网络 主成分分析 数据融合 信息处理 线性估计
Abstract无线传感器网络是集成传感器技术、嵌入式计算技术、无线通信技术以及分布式技术而形成的一种全新的信息获取和处理技术。无线传感器网络将逻辑上的信息世界 和真实的物理世界融合在一起,从而彻底改变人与自然交互的方式。 传感器节点通常部署在大型连续监测系统或感兴趣的区域,由于传感器的密集部署模式,相邻的传感器对同一事件进行监测所获得的数据具有相似性,而传感器节点 在能量、存储空间与计算能力上有限,且节点发送和接受数据的能耗要远大于计算与存储能耗,因此冗余数据的传送在一定程度上将造成网络拥塞和数据冲突以及消 耗过多的能量,降低数据的收集效率,缩短整个网络的生命周期。为避免上述问题,无线传感器网络在从各个传感节点收集数据的过程中,利用传感器节点的本地计 算和存储能力对感知的数据进行融合处理,组合出更高效的、更符合用户需求的准确数据。 本文在主成分分析融合方法的基础上提出一种无线传感器网络簇内分级数据融合算法,先采用自学习加权方法估计出各个传感器的测量方差,通过线性无偏最小方差 估计法对传感器节点的测量数据进行修正,用主成分分析方法得出各传感器的综合支持度和数据融合的公式。应用实例和仿真实验结果验证了该方法的有效性和可靠 性。
Other AbstractWireless sensor network, which is composed of sensor, embedded computing technology, wireless communication and distributed technology, is a kind of novel information acquisition and processing technology. It will integrate logical information world with real physical world and revolutionize the way which people interact with nature. Sensor nodes are usually deployed in large number for continuous monitoring of a system or an area of interest. Because of the dense pattern of sensor deployment, neighboring sensor nodes may sense similar data on a specific phenomenon. while the capability of sensor nodes is restricted in energy, storage and computing, and energy consumption of receiving and sending the data by the node is far greater than that of the computing and storing, so redundant data transmitted will result in congestion of network and data conflict in a certain extent, consume too much energy, reduce efficiency of data collection and minimize the lifetime of the whole network. To avoid these problems, in the process of collecting data from each sensor node of wireless sensor network, data sensed is locally aggregated, processed and combined by computing and storage of each node. Finally, the user is interested in the final data, which is more efficient and accurate. An improved WSN cluster hierarchical data aggregation algorithm is proposed on the basis of the Principal Component Analysis. Firstly, Self-learning weighted method estimates measured variance of every sensor. Secondly, the linear unbiased minimum variance estimate method adopted which is able to reduce the errors of measured datum of the sensor nodes. The formulas of comprehensive support degree of each sensor and data aggregation are obtained according to the PCA method. Finally, the application instance and simulation results verify validity and reliability of the proposed method.
Document Type学位论文
Recommended Citation
GB/T 7714
李海永. 无线传感器网络数据融合技术的研究[D]. 北京. 中国科学院研究生院,2011.
Files in This Item:
File Name/Size DocType Version Access License
无线传感器网络数据融合技术的研究.pdf(1031KB)学位论文 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[李海永]'s Articles
Baidu academic
Similar articles in Baidu academic
[李海永]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[李海永]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 无线传感器网络数据融合技术的研究.pdf
Format: Adobe PDF
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.