|Table of Contents|

Semi infinite Quadratic Optimization Techniques for the Design of Fir Digital Filters(PDF)

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

Issue:
1996年02期
Page:
71-74
Research Field:
Publishing date:

Info

Title:
Semi infinite Quadratic Optimization Techniques for the Design of Fir Digital Filters
Author(s):
Zhi Wenzhu Liu Zhong
School of Electronic Engineering and Optoelectronic Technology, NUST, Nanjing 210094
Keywords:
filter o pt imizat ion methods computer aided design
PACS:
TN713.1
DOI:
-
Abstract:
Parks-McClellan method is a w idely used method for the desig n of FIR f ilters. The method has minimized the max imum error between the desir ed filter and the designed one reg ar dless o f the error energ y. T his paper propo ses a new method fo r the design of FIR f ilter by using semi-infinite quadr at ic opt imizat ion ( SIQO) techniques. T his metho d formulates the FIR filter desig n into SIQO model, w hich minimizes the error energy betw een the desired f ilter and the designed one subject to the specif icat io ns of the peak error . In compar ison w ith the Parks-McClellan method, the new method has g reatly reduced sto pband err or energy w ith the sl ig ht increase of max imum stopband g ain. Design examples show the effect iveness of the new metho d.

References:

1 Hamming R W. Digital filt ers. 3rd ed. New Jersey : Prentice-Hall Eng lew o od, 1989. 127~ 133
2 Par k T W, McClellan J H. Cheby shev appro x imation fo r nonr ecur sive digital filt ers with linear phase. IEEE Tr ans Cir cuit T heo ry , 1972, 19( 3) : 189~194
3 Vaidy anat han P P. Mult ira te digital filt ers, f ilt er banks, polyphase netwo rks, and applications. A Tuto rial Pro c IEEE, 1990, 78( 1) : 56~93
4  Oppenheim A V, Schafer R W. Discrete-time sig nal pr ocessing. New Yo rk: Eng lew o od Chiff s Prentice-Hall, 1989
5  Liu Z, Gong Y H. Semi-infinite quadr atic o ptimizatio n method for the desig n o f r obust a daptiv e ar ray pr ocessor s. IEE Pr o c F, 1990, 137( 3) : 177~ 182

Memo

Memo:
-
Last Update: 2013-04-11