基于可信度估计的迭代滤波的WSNs数据融合算法.pdf

第39卷第4期
电子器件
Vo. 39 No 4
2016年8月
Chinese Journal of Electron Devices
Aug.2016
An Ascertaining Trustworthiness of Sensor Node-based Iterative
Filtering Data Aggregation Algorithm in WSNS
LIU Shuan, SONG Xizhong
College of Information Engineering, Huanghuai University, Zhumadian Henan 463000, China
Abstract: As certaining trustworthiness of data is crucial for WSNS. Iterative filtering(IF)algorithms hold great
promise for such a purpose. However, it is not reasonable to assume that the initial trustworthiness of sensor node is
the same for traditional IF algorithm. Therefore, the ascertaining trustworthiness of sensor node - based iterative fil
tering data aggregation(ATI] DA)algorithm is proposed. In ATIFDA, initial trustworthiness of sensor node is esti-
mated by MLE(Maximum Likelihood Estimation). Firstly, The sensor nodes are divided into disjoint clusters and
each cluster has a cluster head( CH). It collects the reading of sensors, and is used to estimate the initial trustworthi-
ness
by MLE. Secondly, H would cope with these data by iterative filtering. Finally, the data aggregation is done
The simulation results how that the proposed routing al ori m outper orms the K VD algorithm especially in te
Root Mean Squared error(RRSE), which is reduced about 43%
Key words: wireless sensor network data aggregation; cluster; trustworthiness; iterative filtering; maximum likeli
hood estimation
BBACC:6140;:7230
doi:10.3969/j.iss,1005-9490.2016.04.012
基于可信度估计的迭代滤波的WSNs数据融合算法
刘栓',宋喜忠
(黄淮学院信息工程学院,河南驻马店463000)
摘要:传统的迭代滤波IF( Iterative filtering)算法假定传感节点的初始可信度相同,这有背于客观事实。为此,提出基于可
信度估计的迭代滤波的WSNs数据融合 ATIFDA( Ascertaining Trustworthiness of Sensor Node- based Iterative Filtering Data Aggre
gation)算法。 ATIFDA算法利用最大似然估计MLE( Maximum Likelihood Estimation)计算每个传感节点的初始可信度。首先
进行簇划分,每个簇选择一个簇头CH( Cluster head),簇头收集传感节点的感测数据,然后,再利用MILE估计每个传感节点的
初始可信度,再进行F算法,最后,进行数据融合。仿真结果表明,与dKVD算法相比,提出的 ATIFDA算法在噪声环境下,均
方根误差RMSE( Root Mean Squared error)降低了近43%。
关键词:无线传感网;数据融合;簇;可信度;最大似然估计;迭代滤波
中图分类号:TP393
文献标识码:A
文章编号:1005-9490(2016)04-0810-06
为了提高数据传输效率,无线传感网WSNs传感节点数据进行可信度( Trustworthiness)估计。
( Wireless Sensor Networks)的融合节点AN( Aggrega-估计的方差应该接近于CRLB( Cramer Rao Lower
tor Node)将来自多个传感节点提供的数据进行融 bound)。换而言之,方差应逼近于最大似然估计
合,然后再转发到基站( Base Station)。然而,由MLE( Maximum Likelihood Estimator)。尽管传感网
于计算能力和传感节点能量的受限特性,以往数据络应用领域日益广泛,但是对分布式传感节点提供
融合方案仅使用简单的融合算法,如平均算法。但的感测数据进行估计仍是一项挑战性工作。不过
是,简单融合算法很容量受到遭受到恶意攻击的侵随着处理器的功耗减少、计算能力的増加以及成本
害。因此,融合节点AN在数据融合之前需对各自的下降,传感节点能够承载较复杂的融合算法。例
项目来源:河南省科技发展计划项目(132102210463);河南省教育厅科学技术研究重点项目(13A520786)
收稿日期:2015-08-25
修改日期:2015-10-27