This letter develops theoretical analysis of the normalized LMS algorithm (NLMSA) for use in complex-domain adaptive filters in the presence of impulse noise at filter input. We propose a new "stochastic" model for such impulse noise, and assume that filter reference input process is a white process, e.g., digital QAM data, White & Gaussian process, etc. In the analysis, we derive a simple difference equation for mean square tap weight misalignment (MSTWM). Experiment is carried out to demonstrate effectiveness of the NLMSA in robust filtering in the presence of the impulse noise at the filter input. Good agreement between simulated and theoretically calculated filter convergence, in a transient phase as well as in a steady-state, proves the validity of the analysis.
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.
Copy
Shin'ichi KOIKE, "Performance Analysis of the Normalized LMS Algorithm for Complex-Domain Adaptive Filters in the Presence of Impulse Noise at Filter Input" in IEICE TRANSACTIONS on Fundamentals,
vol. E89-A, no. 9, pp. 2422-2428, September 2006, doi: 10.1093/ietfec/e89-a.9.2422.
Abstract: This letter develops theoretical analysis of the normalized LMS algorithm (NLMSA) for use in complex-domain adaptive filters in the presence of impulse noise at filter input. We propose a new "stochastic" model for such impulse noise, and assume that filter reference input process is a white process, e.g., digital QAM data, White & Gaussian process, etc. In the analysis, we derive a simple difference equation for mean square tap weight misalignment (MSTWM). Experiment is carried out to demonstrate effectiveness of the NLMSA in robust filtering in the presence of the impulse noise at the filter input. Good agreement between simulated and theoretically calculated filter convergence, in a transient phase as well as in a steady-state, proves the validity of the analysis.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1093/ietfec/e89-a.9.2422/_p
Copy
@ARTICLE{e89-a_9_2422,
author={Shin'ichi KOIKE, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Performance Analysis of the Normalized LMS Algorithm for Complex-Domain Adaptive Filters in the Presence of Impulse Noise at Filter Input},
year={2006},
volume={E89-A},
number={9},
pages={2422-2428},
abstract={This letter develops theoretical analysis of the normalized LMS algorithm (NLMSA) for use in complex-domain adaptive filters in the presence of impulse noise at filter input. We propose a new "stochastic" model for such impulse noise, and assume that filter reference input process is a white process, e.g., digital QAM data, White & Gaussian process, etc. In the analysis, we derive a simple difference equation for mean square tap weight misalignment (MSTWM). Experiment is carried out to demonstrate effectiveness of the NLMSA in robust filtering in the presence of the impulse noise at the filter input. Good agreement between simulated and theoretically calculated filter convergence, in a transient phase as well as in a steady-state, proves the validity of the analysis.},
keywords={},
doi={10.1093/ietfec/e89-a.9.2422},
ISSN={1745-1337},
month={September},}
Copy
TY - JOUR
TI - Performance Analysis of the Normalized LMS Algorithm for Complex-Domain Adaptive Filters in the Presence of Impulse Noise at Filter Input
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2422
EP - 2428
AU - Shin'ichi KOIKE
PY - 2006
DO - 10.1093/ietfec/e89-a.9.2422
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E89-A
IS - 9
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - September 2006
AB - This letter develops theoretical analysis of the normalized LMS algorithm (NLMSA) for use in complex-domain adaptive filters in the presence of impulse noise at filter input. We propose a new "stochastic" model for such impulse noise, and assume that filter reference input process is a white process, e.g., digital QAM data, White & Gaussian process, etc. In the analysis, we derive a simple difference equation for mean square tap weight misalignment (MSTWM). Experiment is carried out to demonstrate effectiveness of the NLMSA in robust filtering in the presence of the impulse noise at the filter input. Good agreement between simulated and theoretically calculated filter convergence, in a transient phase as well as in a steady-state, proves the validity of the analysis.
ER -