CoChat

A Simple Ladder Realization of Maximally Flat Allpass Fractional Delay Filters

Open access
Koshita, Shunsuke, Abe, Masahide, Kawamata, Masayuki · 2014 · IEEE Transactions on Circuits and Systems II: Express Briefs

The DOI matches its Crossref record; 25 of the 25 references it cites were found in a registry or an index; none of the 20 sources its claims cite is checked for support yet.

Abstract

This brief proposes a new ladder structure for the Thiran fractional delay filter (i.e., maximally flat allpass fractional delay filter given by the Thiran approximation). The proposed ladder structure is based on a continued fraction representation. Although there exists a similar approach that was proposed by Tassart and Depalle, their structure is not realizable because of delay-free loops. On the other hand, we show that the proposed method avoids generating delay-free loops and thus successfully yields a realizable ladder structure for the Thiran fractional delay filter in a very simple form.

Overview

CoChat Summary

TL;DR

This brief proposes a realizable ladder structure for the Thiran maximally flat allpass fractional delay filter using a continued fraction formulation that avoids the delay-free loops of an earlier Tassart–Depalle approach. The resulting structure is simple, stable, and each ladder section requires only two multipliers.

Key findings

  • The proposed ladder structure avoids delay-free loops by using the substitution \(z^{-1} \leftarrow z^{-1}/(1-z^{-1})\), whereas the earlier method of [9] used \(z^{-1} \leftarrow z^{-1}-1\) and produced an unrealizable structure.
  • Each ladder section uses very simple coefficients calculated from \(D\) and \(k\); the integer coefficients \(2\) and \(2k-1\) can be implemented by shifters and adders, so each section requires only two multipliers.
  • Stability is guaranteed: for \(1 \le k \le N\), each ladder section has a single pole \(z=(D-3k+2)/(D+k)\), and the paper proves \(|(D-3k+2)/(D+k)|<1\) for \(D>N-1\), so all poles lie inside the unit circle.
  • The prototype filter before substitution is not necessarily stable and is not allpass; the substitution maps unstable poles inside the unit circle and converts it into the desired stable allpass maximally flat fractional delay filter. The paper also reports experimentally confirming that state variables are mutually orthogonal for white noise input.

Methods

  • Theoretical derivation: the paper introduces a continued fraction expansion involving \(x\) and \(y\), sets \(x=z^{-1}\) and \(y=-D\), and uses the \((2N+1)\)st approximant to obtain a new formulation \(G_{\mathrm{FD}}(z)\) identical to the \(N\)th-order Thiran fractional delay filter \(H_{\mathrm{FD}}(z)\).
  • Structure construction: it first builds a prototype ladder for \(G_p(z)\) by modifying the type IB ladder form of Mitra and Sherwood, then applies substitutions to obtain the proposed ladder structure for the Thiran filter.
  • Analysis: stability is proved by deriving the pole of each ladder section and showing it lies inside the unit circle; no empirical dataset or numerical simulation study is reported.

Limitations

  • The paper states that theoretical analysis of the finite wordlength performance of the proposed ladder structure is beyond its scope.
  • The paper states that the orthogonal property of state variables for white noise input was experimentally confirmed, but theoretical proof and a practical application are still being investigated.
  • Evident from the text: no quantitative implementation results, finite wordlength measurements, or comparisons with other structures are provided.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1109/tcsii.2013.2296131

Cites

25 of 25 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Fig. 1 Basic building blocks for G p ( z ) . (a) Block diagram of P k ( z ) . (b) Block diagram of 1 /Q k ( z ) .
Fig. 1 Basic building blocks for G p ( z ) . (a) Block diagram of P k ( z ) . (b) Block diagram of 1 /Q k ( z ) .From the PDF
Fig. 2 Ladder structure for the prototype filter. (a) Block diagram of G p ( z ) . (b) Block diagram of G pk ( z ) .
Fig. 2 Ladder structure for the prototype filter. (a) Block diagram of G p ( z ) . (b) Block diagram of G pk ( z ) .From the PDF
Fig. 3 Proposed ladder structure for the N th-order Thiran fractional delay filter. (a) block diagram of G FD ( z ) . (b) Block diagram of G k ( z ) .
Fig. 3 Proposed ladder structure for the N th-order Thiran fractional delay filter. (a) block diagram of G FD ( z ) . (b) Block diagram of G k ( z ) .From the PDF

Verification

Claims the paper makes

0 Valid · 0 Not valid · 17 Not checked · 26 Own results and interpretations

28 more claims

Create a free account to see all 43 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 5
  • It may state two separate claims as one. 5

Create a free account to see these claims.

Citations

Cites

Research works printed but not linked: 1

  1. stoyanov, 2011, Digital Filters

Cited by

See all 5

Metadata

Authors
Koshita, S., Abe, M., & Kawamata, M.
Affiliations
Tohoku University (JP)
Venue
IEEE Transactions on Circuits and Systems II: Express Briefs
Published
2014-03
Volume
61
Issue
3
Pages
203-207
ISSN
  • 1549-7747 (ISSN-L)
  • 1558-3791
OpenAlex
W2085274624
Cited by
5
References
23
Citations by year
2023: 1 · 2022: 2 · 2020: 2
FWCI
0.00
Type
journal article
Language
en
Open access
Locations
IEEE Transactions on Circuits & Systems II Express Briefs (publishedVersion)
Funders
Japan Society for the Promotion of Science (24760286)
Topics
  • Physical Sciences › Computer Science › Signal Processing › Digital Filter Design and Implementation
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Analog and Mixed-Signal Circuit Design
  • Physical Sciences › Computer Science › Signal Processing › Advanced Adaptive Filtering Techniques
Keywords
fractional delay filter, all-pass filter, continued fraction representation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.