The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Une architecture pipeline est proposée pour le filtre numérique adaptatif (ADF) normalisé aux moindres carrés moyens (NLMS). La mise en œuvre en pipeline du NLMS n’a pas encore été proposée. L'architecture proposée est la première tentative de mise en œuvre du NLMS ADF de manière pipeline. L'architecture est basée sur une expression équivalente du NLMS dérivée dans cette étude. Il est démontré que l'architecture proposée atteint un chemin critique constant et court sans produire de latence de sortie. De plus, il conserve l'avantage du NLMS, c'est-à-dire que la taille du pas qui assure la convergence est déterminée automatiquement. Des résultats de simulation informatique qui confirment que l'architecture proposée atteint des caractéristiques de convergence identiques à celles du NLMS.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Akio HARADA, Kiyoshi NISHIKAWA, Hitoshi KIYA, "A Pipelined Architecture for Normalized LMS Adaptive Digital Filters" in IEICE TRANSACTIONS on Fundamentals,
vol. E82-A, no. 2, pp. 223-229, February 1999, doi: .
Abstract: A pipelined architecture is proposed for the normalized least mean square (NLMS) adaptive digital filter (ADF). Pipelined implementation of the NLMS has not yet been proposed. The proposed architecture is the first attempt to implement the NLMS ADF in the pipelined fashion. The architecture is based on an equivalent expression of the NLMS derived in this study. It is shown that the proposed architecture achieves a constant and a short critical path without producing output latency. In addition, it retains the advantage of the NLMS, i. e. , that the step size that assures the convergence is determined automatically. Computer simulation results that confirm that the proposed architecture achieves convergence characteristics identical to those of the NLMS.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e82-a_2_223/_p
Copier
@ARTICLE{e82-a_2_223,
author={Akio HARADA, Kiyoshi NISHIKAWA, Hitoshi KIYA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={A Pipelined Architecture for Normalized LMS Adaptive Digital Filters},
year={1999},
volume={E82-A},
number={2},
pages={223-229},
abstract={A pipelined architecture is proposed for the normalized least mean square (NLMS) adaptive digital filter (ADF). Pipelined implementation of the NLMS has not yet been proposed. The proposed architecture is the first attempt to implement the NLMS ADF in the pipelined fashion. The architecture is based on an equivalent expression of the NLMS derived in this study. It is shown that the proposed architecture achieves a constant and a short critical path without producing output latency. In addition, it retains the advantage of the NLMS, i. e. , that the step size that assures the convergence is determined automatically. Computer simulation results that confirm that the proposed architecture achieves convergence characteristics identical to those of the NLMS.},
keywords={},
doi={},
ISSN={},
month={February},}
Copier
TY - JOUR
TI - A Pipelined Architecture for Normalized LMS Adaptive Digital Filters
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 223
EP - 229
AU - Akio HARADA
AU - Kiyoshi NISHIKAWA
AU - Hitoshi KIYA
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E82-A
IS - 2
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - February 1999
AB - A pipelined architecture is proposed for the normalized least mean square (NLMS) adaptive digital filter (ADF). Pipelined implementation of the NLMS has not yet been proposed. The proposed architecture is the first attempt to implement the NLMS ADF in the pipelined fashion. The architecture is based on an equivalent expression of the NLMS derived in this study. It is shown that the proposed architecture achieves a constant and a short critical path without producing output latency. In addition, it retains the advantage of the NLMS, i. e. , that the step size that assures the convergence is determined automatically. Computer simulation results that confirm that the proposed architecture achieves convergence characteristics identical to those of the NLMS.
ER -