Adaptive filtering under maximum mutual information criterion

Badong Chen; Jinchun Hu; Hongbo Li; Zengqi Sun

doi:10.1016/j.neucom.2008.02.003

Adaptive filtering under maximum mutual information criterion

Badong Chen
, Jinchun Hu
, Hongbo Li
, Zengqi Sun

Tsinghua University

科研成果: 期刊稿件 › 文章 › 同行评审

9 引用（Scopus）

摘要

The maximum mutual information (MaxMI) criterion is used as the adaptation cost for the adaptive filtering. This criterion is robust to measure distortions, and has strong connection with traditional mean-square error (MSE) criterion. Under Gaussian assumption, the closed-form solution of the finite impulse response (FIR) filter is obtained. Further, based on the kernel density estimation, the stochastic mutual information gradient (SMIG) algorithm is derived. Simulation results emphasize the robustness of this new algorithm.

源语言	英语
页（从-至）	3680-3684
页数	5
期刊	Neurocomputing
卷	71
期	16-18
DOI	https://doi.org/10.1016/j.neucom.2008.02.003
出版状态	已出版 - 10月 2008
已对外发布	是

访问文件

10.1016/j.neucom.2008.02.003

其它文件与链接

链接到 Scopus 的出版物

引用此

@article{8cb6bafc3552463381b49cb8e26d9c51,

title = "Adaptive filtering under maximum mutual information criterion",

abstract = "The maximum mutual information (MaxMI) criterion is used as the adaptation cost for the adaptive filtering. This criterion is robust to measure distortions, and has strong connection with traditional mean-square error (MSE) criterion. Under Gaussian assumption, the closed-form solution of the finite impulse response (FIR) filter is obtained. Further, based on the kernel density estimation, the stochastic mutual information gradient (SMIG) algorithm is derived. Simulation results emphasize the robustness of this new algorithm.",

keywords = "Adaptive filtering, Mutual information, Stochastic gradient algorithm, System identification",

author = "Badong Chen and Jinchun Hu and Hongbo Li and Zengqi Sun",

year = "2008",

month = oct,

doi = "10.1016/j.neucom.2008.02.003",

language = "英语",

volume = "71",

pages = "3680--3684",

journal = "Neurocomputing",

issn = "0925-2312",

publisher = "Elsevier B.V.",

number = "16-18",

}

TY - JOUR

T1 - Adaptive filtering under maximum mutual information criterion

AU - Chen, Badong

AU - Hu, Jinchun

AU - Li, Hongbo

AU - Sun, Zengqi

PY - 2008/10

Y1 - 2008/10

N2 - The maximum mutual information (MaxMI) criterion is used as the adaptation cost for the adaptive filtering. This criterion is robust to measure distortions, and has strong connection with traditional mean-square error (MSE) criterion. Under Gaussian assumption, the closed-form solution of the finite impulse response (FIR) filter is obtained. Further, based on the kernel density estimation, the stochastic mutual information gradient (SMIG) algorithm is derived. Simulation results emphasize the robustness of this new algorithm.

AB - The maximum mutual information (MaxMI) criterion is used as the adaptation cost for the adaptive filtering. This criterion is robust to measure distortions, and has strong connection with traditional mean-square error (MSE) criterion. Under Gaussian assumption, the closed-form solution of the finite impulse response (FIR) filter is obtained. Further, based on the kernel density estimation, the stochastic mutual information gradient (SMIG) algorithm is derived. Simulation results emphasize the robustness of this new algorithm.

KW - Adaptive filtering

KW - Mutual information

KW - Stochastic gradient algorithm

KW - System identification

UR - https://www.scopus.com/pages/publications/56549090787

U2 - 10.1016/j.neucom.2008.02.003

DO - 10.1016/j.neucom.2008.02.003

M3 - 文章

AN - SCOPUS:56549090787

SN - 0925-2312

VL - 71

SP - 3680

EP - 3684

JO - Neurocomputing

JF - Neurocomputing

IS - 16-18

ER -

Adaptive filtering under maximum mutual information criterion

摘要

访问文件

其它文件与链接

学术指纹

引用此