Non-Uniform Norm Constraint LMS Algorithm for Sparse System Identification

Non-Uniform Norm Constraint LMS Algorithm for Sparse System Identification
复制标题

稀疏系统辨识的非均匀范数约束LMS算法

DOI:
10.1109/lcomm.2013.011113.121586
复制
发表时间:
2013-01
期刊:
IEEE Communications Letters
影响因子:
--
通讯作者:
Tong, F.
Tong, F.
中科院分区:
其他
文献类型:
--
作者:
Wu, F. Y.;Tong, F.

文献摘要

参考文献

被引文献

相似文献

稀疏性一直被用来改善基于最小均方(LMS)的稀疏系统辨识的性能,其形式为l0范数或l1范数约束。然而,对于具有不同稀疏度的特定系统的最优范数约束,缺乏理论研究。本文提出了一种在范数约束的稀疏性利用效应与其产生的估计偏差之间寻求折衷的方法,由此导出了一种新的算法,该算法对经典LMS算法的代价函数进行了非均匀范数(类p范数)惩罚。这种修改等效于根据所有条目中每个滤波器系数的相对值在迭代上施加l0范数或l1范数零吸引元素的序列。数值仿真结果表明,该方法具有收敛速度快、对不同稀疏性具有较好的容错性等优点。
Sparsity property has long been exploited to improve the performance of least mean square (LMS) based identification of sparse systems, in the form of l0-norm or l1-norm constraint. However, there is a lack of theoretical investigations regarding the optimum norm constraint for specific system with different sparsity. This paper presents an approach by seeking the tradeoff between the sparsity exploitation effect of norm constraint and the estimation bias it produces, from which a novel algorithm is derived to modify the cost function of classic LMS algorithm with a non-uniform norm (p-norm like) penalty. This modification is equivalent to impose a sequence of l0-norm or l1-norm zero attraction elements on the iteration according to the relative value of each filter coefficient among all the entries. The superiorities of the proposed method including improved convergence rate as well as better tolerance upon different sparsity are demonstrated by numerical simulations.
DOI: 10.1109/jstsp.2009.2039173
发表时间: 2010-04-01
影响因子: 7.5
作者:
Jin, Jian;Gu, Yuantao;Mei, Shunliang
通讯作者: Mei, Shunliang
DOI: 10.1109/lsp.2009.2024736
发表时间: 2009-09-01
影响因子: 3.9
作者:
Gu, Yuantao;Jin, Jian;Mei, Shunliang
通讯作者: Mei, Shunliang
DOI: 10.1017/cbo9780511794308
发表时间: 2012
期刊: --
影响因子: --
作者:
Gitta Kutyniok
通讯作者: Gitta Kutyniok
DOI: 10.1109/lsp.2007.898300
发表时间: 2007-10-01
影响因子: 3.9
作者:
Chartrand, Rick
通讯作者: Chartrand, Rick
DOI: 10.1109/78.738251
发表时间: 1999-01-01
影响因子: 5.4
作者:
Rao, BD;Kreutz-Delgado, K
通讯作者: Kreutz-Delgado, K