无线传感器网络基于压缩感知数据处理算法研究  

WSN based on distributed compressed sensing data processing algorithms

作　　者：庄严[1] 沈军[2] 庄田 

机构地区：[1]桂林电子科技大学计算机科学与工程学院,广西桂林541004 [2]桂林空军学院科研部,广西桂林541003 [3]杭州天丽科技有限公司科研部,浙江杭州310012

出　　处：《计算机工程与设计》2013年第9期3028-3032,共5页Computer Engineering and Design

摘　　要：无线传感器网络中的分布式数据处理处理时,会出现数据压缩率低和数据相关性差的特点,在数据采集和处理中采用分布式压缩感知算法。对基础矩阵和观测矩阵的合理建立,在满足无线传感器网络数据传输的可靠性同时,充分考虑数据之间的相关性,大大降低数据的冗余度,减少了数据量的传输。实验结果表明,在满足信号数据可靠性的同时,有效的增加了信号的压缩比,减少了节点数据采集量。In view of low data compression rate and poor data correlation, which is caused by distributed data storage in the wire- less sensor, compression perception algorithm is adopted in data collection and treatment. According to an analysis of the basis and observation matrix, a reasonable choice is made, which meets the need of reliability of data transmission in wireless sensor network. In the meantime, with adequate consideration of the correlation among data, the data redundancy and the amount of data transmission are greatly reduled. And the results show it effectively increases the signal compression ratio and decreases the number of nodes. Moreover, the reliability of data signal is met simultaneously.

关 键 词：无线传感器网络 压缩感知 路由 均方误差 稀疏矩阵 

分 类 号：TN915[电子电信—通信与信息系统]