Least absolute deviation neural network and its application to time delay estimation

Zhishun Wang; J. Y. Cheung; Y. S. Xia; J. D.Z. Chen

Least absolute deviation neural network and its application to time delay estimation

Zhishun Wang, J. Y. Cheung, Y. S. Xia, J. D.Z. Chen

Research output: Contribution to conference › Paper › peer-review

Abstract

Least absolute deviation (LAD) optimization model, or called unconstrained minimum L₁-norm optimization model, has found extensive applications in linear parameter estimations. In this paper, neural implementation of LAD optimization model is presented, where a new neural network is constructed and its performance in performing LAD optimization is evaluated theoretically and experimentally. Then, the application of the proposed LAD network to time delay estimation (TDE) is presented. Compared with the popular time-delay estimation methods based on higher order spectra (HOS), our method is free of the assumption that the signal is non-Gaussian and the noises are Gaussian, which is closer to real situations.

Original language	English (US)
Pages	13-18
Number of pages	6
State	Published - 1999
Externally published	Yes
Event	Proceedings of the 1999 Artificial Neural Networks in Engineering Conference (ANNIE '99) - St. Louis, MO, USA Duration: Nov 7 1999 → Nov 10 1999

Other

Other	Proceedings of the 1999 Artificial Neural Networks in Engineering Conference (ANNIE '99)
City	St. Louis, MO, USA
Period	11/7/99 → 11/10/99

ASJC Scopus subject areas

Software

Cite this

@conference{3e4854956c0548baa8dc39d25c1731ee,

title = "Least absolute deviation neural network and its application to time delay estimation",

abstract = "Least absolute deviation (LAD) optimization model, or called unconstrained minimum L1-norm optimization model, has found extensive applications in linear parameter estimations. In this paper, neural implementation of LAD optimization model is presented, where a new neural network is constructed and its performance in performing LAD optimization is evaluated theoretically and experimentally. Then, the application of the proposed LAD network to time delay estimation (TDE) is presented. Compared with the popular time-delay estimation methods based on higher order spectra (HOS), our method is free of the assumption that the signal is non-Gaussian and the noises are Gaussian, which is closer to real situations.",

author = "Zhishun Wang and Cheung, {J. Y.} and Xia, {Y. S.} and Chen, {J. D.Z.}",

year = "1999",

language = "English (US)",

pages = "13--18",

note = "Proceedings of the 1999 Artificial Neural Networks in Engineering Conference (ANNIE '99) ; Conference date: 07-11-1999 Through 10-11-1999",

}

TY - CONF

T1 - Least absolute deviation neural network and its application to time delay estimation

AU - Wang, Zhishun

AU - Cheung, J. Y.

AU - Xia, Y. S.

AU - Chen, J. D.Z.

PY - 1999

Y1 - 1999

N2 - Least absolute deviation (LAD) optimization model, or called unconstrained minimum L1-norm optimization model, has found extensive applications in linear parameter estimations. In this paper, neural implementation of LAD optimization model is presented, where a new neural network is constructed and its performance in performing LAD optimization is evaluated theoretically and experimentally. Then, the application of the proposed LAD network to time delay estimation (TDE) is presented. Compared with the popular time-delay estimation methods based on higher order spectra (HOS), our method is free of the assumption that the signal is non-Gaussian and the noises are Gaussian, which is closer to real situations.

AB - Least absolute deviation (LAD) optimization model, or called unconstrained minimum L1-norm optimization model, has found extensive applications in linear parameter estimations. In this paper, neural implementation of LAD optimization model is presented, where a new neural network is constructed and its performance in performing LAD optimization is evaluated theoretically and experimentally. Then, the application of the proposed LAD network to time delay estimation (TDE) is presented. Compared with the popular time-delay estimation methods based on higher order spectra (HOS), our method is free of the assumption that the signal is non-Gaussian and the noises are Gaussian, which is closer to real situations.

UR - http://www.scopus.com/inward/record.url?scp=0033345272&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033345272&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0033345272

SP - 13

EP - 18

T2 - Proceedings of the 1999 Artificial Neural Networks in Engineering Conference (ANNIE '99)

Y2 - 7 November 1999 through 10 November 1999

ER -

Least absolute deviation neural network and its application to time delay estimation

Abstract

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this