Simulation of fixed-point arithmetic in the lms and vs adaptive filter algorithms

A. Mämmelä; I. Saarinen

doi:10.1109/ISSSTA.1992.665699

Simulation of fixed-point arithmetic in the lms and vs adaptive filter algorithms

Not published at VTT

University of Oulu

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

Abstract

Fixed-point two's-complement arithmetic was simulated in the LMS and the VS algorithms. The simulation method is first briefly described. The algorithms were used as interference suppression algorithms in a DS spread-spectrum system. The simulated performance results were compared to the theoretical results. It was found that the theoretical results are optimistic in the present application where the covariance matrix of the input signal is often ill conditioned. The VS algorithm is more robust than the LMS algorithm against the finite word-length effects since the former has the freedom to select the best possible step size at different word lengths.

Original language	English
Title of host publication	IEEE Second International Symposium on Spread Spectrum Techniques and Applications
Publisher	IEEE Institute of Electrical and Electronic Engineers
Pages	271-274
DOIs	https://doi.org/10.1109/ISSSTA.1992.665699
Publication status	Published - 1992
MoE publication type	A4 Article in a conference publication
Event	2nd IEEE International Symposium on Spread Spectrum Techniques and Applications, ISSTA 1992 - Yokohama, Japan Duration: 29 Nov 1992 → 2 Dec 1992

Conference

Conference	2nd IEEE International Symposium on Spread Spectrum Techniques and Applications, ISSTA 1992
Country/Territory	Japan
City	Yokohama
Period	29/11/92 → 2/12/92

Access to Document

10.1109/ISSSTA.1992.665699

Cite this

@inproceedings{8a3c0660b7c548888dc59fb6e56922cb,

title = "Simulation of fixed-point arithmetic in the lms and vs adaptive filter algorithms",

abstract = "Fixed-point two's-complement arithmetic was simulated in the LMS and the VS algorithms. The simulation method is first briefly described. The algorithms were used as interference suppression algorithms in a DS spread-spectrum system. The simulated performance results were compared to the theoretical results. It was found that the theoretical results are optimistic in the present application where the covariance matrix of the input signal is often ill conditioned. The VS algorithm is more robust than the LMS algorithm against the finite word-length effects since the former has the freedom to select the best possible step size at different word lengths.",

author = "A. M{\"a}mmel{\"a} and I. Saarinen",

note = "Publisher Copyright: {\textcopyright} 1991 Institute of Electrical and Electronics Engineers Inc. All rights reserved.; 2nd IEEE International Symposium on Spread Spectrum Techniques and Applications, ISSTA 1992 ; Conference date: 29-11-1992 Through 02-12-1992",

year = "1992",

doi = "10.1109/ISSSTA.1992.665699",

language = "English",

pages = "271--274",

booktitle = "IEEE Second International Symposium on Spread Spectrum Techniques and Applications",

publisher = "IEEE Institute of Electrical and Electronic Engineers",

address = "United States",

}

Mämmelä, A & Saarinen, I 1992, Simulation of fixed-point arithmetic in the lms and vs adaptive filter algorithms. in IEEE Second International Symposium on Spread Spectrum Techniques and Applications. IEEE Institute of Electrical and Electronic Engineers, pp. 271-274, 2nd IEEE International Symposium on Spread Spectrum Techniques and Applications, ISSTA 1992, Yokohama, Japan, 29/11/92. https://doi.org/10.1109/ISSSTA.1992.665699

Simulation of fixed-point arithmetic in the lms and vs adaptive filter algorithms. / Mämmelä, A.; Saarinen, I.
IEEE Second International Symposium on Spread Spectrum Techniques and Applications. IEEE Institute of Electrical and Electronic Engineers, 1992. p. 271-274.

Research output: Chapter in Book/Report/Conference proceeding › Conference article in proceedings › Scientific › peer-review

TY - GEN

T1 - Simulation of fixed-point arithmetic in the lms and vs adaptive filter algorithms

AU - Mämmelä, A.

AU - Saarinen, I.

PY - 1992

Y1 - 1992

N2 - Fixed-point two's-complement arithmetic was simulated in the LMS and the VS algorithms. The simulation method is first briefly described. The algorithms were used as interference suppression algorithms in a DS spread-spectrum system. The simulated performance results were compared to the theoretical results. It was found that the theoretical results are optimistic in the present application where the covariance matrix of the input signal is often ill conditioned. The VS algorithm is more robust than the LMS algorithm against the finite word-length effects since the former has the freedom to select the best possible step size at different word lengths.

AB - Fixed-point two's-complement arithmetic was simulated in the LMS and the VS algorithms. The simulation method is first briefly described. The algorithms were used as interference suppression algorithms in a DS spread-spectrum system. The simulated performance results were compared to the theoretical results. It was found that the theoretical results are optimistic in the present application where the covariance matrix of the input signal is often ill conditioned. The VS algorithm is more robust than the LMS algorithm against the finite word-length effects since the former has the freedom to select the best possible step size at different word lengths.

UR - http://www.scopus.com/inward/record.url?scp=85067832210&partnerID=8YFLogxK

U2 - 10.1109/ISSSTA.1992.665699

DO - 10.1109/ISSSTA.1992.665699

M3 - Conference article in proceedings

AN - SCOPUS:85067832210

SP - 271

EP - 274

BT - IEEE Second International Symposium on Spread Spectrum Techniques and Applications

PB - IEEE Institute of Electrical and Electronic Engineers

T2 - 2nd IEEE International Symposium on Spread Spectrum Techniques and Applications, ISSTA 1992

Y2 - 29 November 1992 through 2 December 1992

ER -

Simulation of fixed-point arithmetic in the lms and vs adaptive filter algorithms

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this