Transient analysis of zero attracting NLMS algorithm without Gaussian inputs assumption
Transient analysis of zero attracting NLMS algorithm without Gaussian inputs assumption
复制标题
DOI：
10.1016/j.sigpro.2013.10.022
复制
发表时间：
2014-04
期刊：
影响因子：
--
通讯作者：
Sheng Zhang;Jiashu Zhang
中科院分区：
文献类型：
--
作者：
Sheng Zhang;Jiashu Zhang
在稀疏系统辨识中，零吸引归一化最小均方（ZA-NLMS）算法比归一化最小均方（NLMS）算法具有更小的稳态误差。对于几种版本的零吸引最小均方算法，大多数可用的分析结果都假设了白高斯输入。本文对无高斯输入假设的ZA-NLMS算法进行了个体权值误差方差分析。IWV分析基于精确的个体权重误差关系，并用于推导ZA-NLMS算法的瞬态和稳态行为，而不限制输入为高斯或白色，同时在评估某些期望时引入一些假设以克服权重非线性。大量的仿真用于验证所提出的分析结果。
The zero attracting normalized least mean square (ZA-NLMS) algorithm achieves lower steady-state error than the normalized least mean square (NLMS) algorithm for sparse system identification. Most of the available analytical results on several versions of the zero attracting least mean square algorithms assume white Gaussian inputs. This paper presents the individual weight error variance (IWV) analysis of the ZA-NLMS algorithm without Gaussian inputs assumption. The IWV analysis is based on exact individual weight error relation and used to derive the transient and steady-state behavior of the ZA-NLMS algorithm without restricting the input to being Gaussian or white, whereas some assumptions are introduced to overcome weight nonlinearity in evaluating certain expectations involved. Extensive simulations are used to verify the analysis results presented.