摘要
针对在传感器网络的参数估计中,测量环境可能存在冲击噪声或脉冲干扰,导致测量值中包含大大偏离实际范围的离群值,从而无法获得有效的参数估计问题,提出了一种基于节点协作、具有顽健性的分布式RLS估计算法,该算法利用在一段时间内,节点受到干扰或攻击通常具有偶发性(即稀疏性)的特点,在代价函数中引入1范数,对测量数据中可能存在的离群值进行识别和剔除,同时利用网络各节点之间的相互协作,进一步提高参数估计的性能。通过计算机仿真实验,验证了该算法具有良好的估计性能。
In the parameter estimation of the sensor network, there may be impact noise or pulse interference in the measurement environment, which leads to the outliers that greatly deviate from the actual range, so that the effective parameter estimation problem can't be obtained. A distributed RLS estimation algorithm based on node cooperation and robustness was proposed. The characteristic of interference or attack on a node is usually sporadic(ie, sparse) for a period of time was utilized, 1 norm was introduced to cost function. The outliers which may exist in the measurement data were identified and eliminated, and the interaction between the nodes of the network was utilized to further improve the performance of the parameter estimation. The computer simulation experiments show that the algorithm has good estimation performance.
引文
[1]孙利民,李建中,陈渝,等.无线传感器网络[M].北京:清华大学出版社,2005.SUN L M,LI J Z,CHEN Y,et al.Wireless sensor network[M].Beijing:Tsinghua University Press,2005.
[2]MA L,WANG Z,LAM H,et al.Distributed event-based set-membership filtering for a class of nonlinear systems with sensor saturations over sensor networks[J].IEEE Transactions on Signal Processing,2017,47(11):3892-3905.
[3]钱萍,吴蒙.无线传感器网络隐私保护方法[J].电信科学,2013,29(1):23-30.QIAN P,WU M.A privacy preserving method in WSN[J].Telecommunications Science,2013,29(1):23-30.
[4]孟凯露,岳克强,尚俊娜.基于元胞蝙蝠算法的无线传感器网络节点定位研究[J].电信科学,2017,33(11):56-65.MENG K L,YUE K Q,SHANG J N.Wireless sensor network nodes localization method based on cellular automata bat algorithm[J].Telecommunications Science,2017,33(11):56-65.
[5]盛积饶.基于分布式递归最小二乘算法的网络稀疏信号处理研究[D].南京:南京理工大学,2016.SHENG J R.Research on network sparse signal processing based on distributed recursive least squares algorithm[D].Nanjing:Nanjing University of Science and Technology,2016.
[6]马兰申.自适应网络的分布式估计研究[D].苏州:苏州大学,2014.MA L S.Research on distributed estimation of adaptive network[D].Suzhou:Soochow University,2014.
[7]LIU Z T,LIU Y,LI C G.Distributed sparse recursive least-squares over networks[J].IEEE Transactions on Signal Processing,2014,62(6):1386-1395.
[8]LIU Y,LI C,ZHANG Z.Diffusion sparse least-mean squares over networks[J].IEEE Transactions on Signal Processing,2012,60(8):4480-4485.
[9]LORENZO P D,SAYED A H.Sparse distributed learning based on diffusion adaptation[J].IEEE Transactions on Signal Processing,2013,61(6):1419-1433.
[10]TAKAHASHI N,YAMADA I,SAYED A H.Diffusion least-mean squares with adaptive combiners-formulation and performance analysis[J].IEEE Transactions on Signal Processing,2010,58(9):4795-4810.
[11]LIU Z T,LI C G,LIU Y G.Distributed censored regression over networks[J].IEEE Transactions on Signal Processing,2015,63(20):5437-5449.
[12]ZAYYANI H,KORKI M,MARVASTI F.A distributed 1-bit compressed sensing algorithm robust to impulsive noise[J].IEEE Communications Letters,2016,20(6):1132-1135.
[13]CHEN P,RONG Y,NORDHOLM S,et al.Joint channel estimation and impulsive noise mitigation in underwater acoustic OFDM communication systems[J].IEEE Transactions on Wireless Communications,2017,16(9):6165-6178.
[14]CATTIVELLI F S,LOPES C G,SAYED A H.Diffusion recursive least-squares for distributed estimation over adaptive networks[J].IEEE Transactions on Signal Processing,2008,56(5):1865-1877.
[15]ANGELOSANTE D,BAZERQUE J A,GIANNAKIS G B.Online adaptive estimation of sparse signals:where RLS meets the 1norm[J].IEEE Transactions on Signals Processing,2010,58(7):3436-3447.
[16]BABADI B,KALOUPTSIDIS N,TAROKH V.SPARLS:the sparse RLS algorithm[J].IEEE Transactions on Signal Processing,2010,58(8):4013-4025.
[17]康凯凯,刘兆霆.传感器网络分布式顽健自适应估计算法[J].传感技术学报,2018,31(4):602-606.KANG K K,LIU Z T.Distributed robust adaptive estimation algorithm for sensor networks[J].Chinese Journal of Sensors and Actuators,2018,31(4):602-606.