Design of maximally decimated near-perfect-reconstruction DFT filter banks with allpass-based analysis filters

Enisa Galijašević; Jörg Kliwer

doi:10.1109/acssc.2001.986989

Design of maximally decimated near-perfect-reconstruction DFT filter banks with allpass-based analysis filters

Enisa Galijašević, Jörg Kliwer

Research output: Contribution to journal › Conference article › peer-review

Abstract

In this paper we address a maximally decimated IIR DFT filter bank, where all polyphase components are replaced by allpass filters. Generally, a perfect reconstruction (PR) solution leads to a synthesis filter bank with unstable subband filters. As a novelty we show that by using closed-form FIR phase compensation filters as polyphase components in the synthesis filter bank we obtain a stable near-PR analysis-synthesis system. Furthermore, the remaining linear and aliasing distortions only depend on the overall system delay, and can be made arbitrarily small.

Original language	English (US)
Pages (from-to)	577-581
Number of pages	5
Journal	Conference Record of the Asilomar Conference on Signals, Systems and Computers
Volume	1
DOIs	https://doi.org/10.1109/acssc.2001.986989
State	Published - 2001
Externally published	Yes
Event	35th Asilomar Conference on Signals, Systems and Computers - Pacific Grove, CA, United States Duration: Nov 4 2001 → Nov 7 2001

All Science Journal Classification (ASJC) codes

Signal Processing
Computer Networks and Communications

Access to Document

10.1109/acssc.2001.986989

Cite this

@article{b08ba4b15bc34e88a3a9ba5d7a20c86f,

title = "Design of maximally decimated near-perfect-reconstruction DFT filter banks with allpass-based analysis filters",

abstract = "In this paper we address a maximally decimated IIR DFT filter bank, where all polyphase components are replaced by allpass filters. Generally, a perfect reconstruction (PR) solution leads to a synthesis filter bank with unstable subband filters. As a novelty we show that by using closed-form FIR phase compensation filters as polyphase components in the synthesis filter bank we obtain a stable near-PR analysis-synthesis system. Furthermore, the remaining linear and aliasing distortions only depend on the overall system delay, and can be made arbitrarily small.",

author = "Enisa Galija{\v s}evi{\'c} and J{\"o}rg Kliwer",

year = "2001",

doi = "10.1109/acssc.2001.986989",

language = "English (US)",

volume = "1",

pages = "577--581",

journal = "Conference Record of the Asilomar Conference on Signals, Systems and Computers",

issn = "1058-6393",

publisher = "IEEE Computer Society",

note = "35th Asilomar Conference on Signals, Systems and Computers ; Conference date: 04-11-2001 Through 07-11-2001",

}

TY - JOUR

T1 - Design of maximally decimated near-perfect-reconstruction DFT filter banks with allpass-based analysis filters

AU - Galijašević, Enisa

AU - Kliwer, Jörg

PY - 2001

Y1 - 2001

N2 - In this paper we address a maximally decimated IIR DFT filter bank, where all polyphase components are replaced by allpass filters. Generally, a perfect reconstruction (PR) solution leads to a synthesis filter bank with unstable subband filters. As a novelty we show that by using closed-form FIR phase compensation filters as polyphase components in the synthesis filter bank we obtain a stable near-PR analysis-synthesis system. Furthermore, the remaining linear and aliasing distortions only depend on the overall system delay, and can be made arbitrarily small.

AB - In this paper we address a maximally decimated IIR DFT filter bank, where all polyphase components are replaced by allpass filters. Generally, a perfect reconstruction (PR) solution leads to a synthesis filter bank with unstable subband filters. As a novelty we show that by using closed-form FIR phase compensation filters as polyphase components in the synthesis filter bank we obtain a stable near-PR analysis-synthesis system. Furthermore, the remaining linear and aliasing distortions only depend on the overall system delay, and can be made arbitrarily small.

UR - http://www.scopus.com/inward/record.url?scp=0035573976&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0035573976&partnerID=8YFLogxK

U2 - 10.1109/acssc.2001.986989

DO - 10.1109/acssc.2001.986989

M3 - Conference article

AN - SCOPUS:0035573976

SN - 1058-6393

VL - 1

SP - 577

EP - 581

JO - Conference Record of the Asilomar Conference on Signals, Systems and Computers

JF - Conference Record of the Asilomar Conference on Signals, Systems and Computers

T2 - 35th Asilomar Conference on Signals, Systems and Computers

Y2 - 4 November 2001 through 7 November 2001

ER -

Design of maximally decimated near-perfect-reconstruction DFT filter banks with allpass-based analysis filters

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this