On the performance of too short adaptive FIR filters

D.W.E. Schobben; P.C.W. Sommen

On the performance of too short adaptive FIR filters

D.W.E. Schobben, P.C.W. Sommen

Signal Processing Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

61 Downloads (Pure)

Abstract

Performance analyses of adaptive algorithms such as LMS and RLS often rely on the assumption that the input signal is stationary. Also it often is assumed that the adaptive nite impulse response FIR filter is long enough to make a good approximation of the unknown system. In many practical situations these assumptions do not hold and the instantaneous misadjustment of the adaptive filter can grow large. In this paper this effect is investigated and two methods are presented to improve the performance of the adaptive filter.

Original language	English
Title of host publication	Proc. CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing
Editors	J.P. Veen
Place of Publication	Utrecht, Netherlands
Publisher	STW Technology Foundation
Pages	545-549
ISBN (Print)	90-73461-12-X
Publication status	Published - 1997
Event	CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing - Mierlo, Netherlands Duration: 27 Nov 1997 → 28 Nov 1997

Conference

Conference	CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing
Abbreviated title	CSSP-97
Country/Territory	Netherlands
City	Mierlo
Period	27/11/97 → 28/11/97

Access to Document

600347147397397Final author version , 162 KB

Cite this

@inproceedings{41af3d661cd6469da737ed820c64f709,

title = "On the performance of too short adaptive FIR filters",

abstract = "Performance analyses of adaptive algorithms such as LMS and RLS often rely on the assumption that the input signal is stationary. Also it often is assumed that the adaptive nite impulse response FIR filter is long enough to make a good approximation of the unknown system. In many practical situations these assumptions do not hold and the instantaneous misadjustment of the adaptive filter can grow large. In this paper this effect is investigated and two methods are presented to improve the performance of the adaptive filter.",

author = "D.W.E. Schobben and P.C.W. Sommen",

year = "1997",

language = "English",

isbn = "90-73461-12-X",

pages = "545--549",

editor = "J.P. Veen",

booktitle = "Proc. CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing",

publisher = "STW Technology Foundation",

note = "CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing, CSSP-97 ; Conference date: 27-11-1997 Through 28-11-1997",

}

Schobben, DWE & Sommen, PCW 1997, On the performance of too short adaptive FIR filters. in JP Veen (ed.), Proc. CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing. STW Technology Foundation, Utrecht, Netherlands, pp. 545-549, CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing, Mierlo, Netherlands, 27/11/97.

On the performance of too short adaptive FIR filters. / Schobben, D.W.E.; Sommen, P.C.W.
Proc. CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing. ed. / J.P. Veen. Utrecht, Netherlands: STW Technology Foundation, 1997. p. 545-549.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - On the performance of too short adaptive FIR filters

AU - Schobben, D.W.E.

AU - Sommen, P.C.W.

PY - 1997

Y1 - 1997

N2 - Performance analyses of adaptive algorithms such as LMS and RLS often rely on the assumption that the input signal is stationary. Also it often is assumed that the adaptive nite impulse response FIR filter is long enough to make a good approximation of the unknown system. In many practical situations these assumptions do not hold and the instantaneous misadjustment of the adaptive filter can grow large. In this paper this effect is investigated and two methods are presented to improve the performance of the adaptive filter.

AB - Performance analyses of adaptive algorithms such as LMS and RLS often rely on the assumption that the input signal is stationary. Also it often is assumed that the adaptive nite impulse response FIR filter is long enough to make a good approximation of the unknown system. In many practical situations these assumptions do not hold and the instantaneous misadjustment of the adaptive filter can grow large. In this paper this effect is investigated and two methods are presented to improve the performance of the adaptive filter.

M3 - Conference contribution

SN - 90-73461-12-X

SP - 545

EP - 549

BT - Proc. CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing

A2 - Veen, J.P.

PB - STW Technology Foundation

CY - Utrecht, Netherlands

T2 - CSSP-97, 8th Annual ProRISC/IEEE Workshop on Circuits, Systems and Signal Processing

Y2 - 27 November 1997 through 28 November 1997

ER -

On the performance of too short adaptive FIR filters

Abstract

Conference

Access to Document

Fingerprint

Cite this