pith. sign in

arxiv: 2604.07680 · v1 · submitted 2026-04-09 · 📡 eess.SP

Low-complexity Frequency Domain Equalization for filtered-AFDM over General Physical Channels

Cheng Shen , Chenyang Zhang , Jinhong Yuan This is my paper

Pith reviewed 2026-05-10 18:24 UTC · model grok-4.3

classification 📡 eess.SP
keywords AFDMfiltered-AFDMfrequency domain equalizationlow-complexity equalizationoff-grid effectshigh-mobility communicationsLMMSE equalizationwideband channels
0
0 comments X p. Extension

The pith

A two-stage frequency domain equalizer for filtered-AFDM approaches full LMMSE performance with far lower complexity.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper examines equalization for affine frequency division multiplexing under realistic general physical channels that include off-grid delay and Doppler shifts. It shows that these shifts produce strong inter-symbol interference when processing occurs in the discrete affine Fourier transform domain. The authors then observe that the channel matrix in the ordinary frequency domain stays compact for wideband systems and use that property to design a two-stage frequency-domain equalizer. Numerical tests confirm the new scheme reaches performance nearly identical to the optimal full-block linear minimum mean square error solution while cutting complexity and beating conventional time-domain methods in wideband cases.

Core claim

Off-grid delay and Doppler effects in general physical channels create severe inter-symbol interference that defeats effective DAFT-domain equalization for filtered-AFDM, yet the frequency-domain channel matrix remains compact in wideband systems; this compactness supports a low-complexity two-stage frequency-domain equalization scheme whose bit-error-rate performance approaches that of full-block LMMSE equalization and exceeds time-domain equalization in wideband high-mobility scenarios.

What carries the argument

The two-stage frequency domain equalization scheme that exploits compactness of the frequency-domain channel matrix to handle off-grid-induced interference.

If this is right

  • The scheme cuts computational complexity relative to full-block LMMSE while retaining near-optimal error performance.
  • It delivers clear gains over time-domain equalization once bandwidth becomes large.
  • It makes filtered-AFDM equalization practical for general physical channels that contain off-grid effects.
  • The approach directly supports deployment of AFDM waveforms in high-mobility wideband links.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar two-stage frequency-domain processing may apply to other multicarrier formats that suffer from off-grid parameter mismatch.
  • In narrower-band or extremely high-mobility regimes the compactness property could weaken, pointing to possible hybrid time-frequency methods as a next step.
  • The technique could combine with existing channel-estimation algorithms to create end-to-end low-complexity receivers for high-mobility systems.

Load-bearing premise

The frequency-domain channel matrix stays compact enough in wideband systems even after off-grid delay and Doppler effects appear.

What would settle it

A simulation or over-the-air test in which the proposed equalizer's error rate deviates markedly from full-block LMMSE under wideband channel conditions that include pronounced off-grid delay and Doppler shifts.

Figures

Figures reproduced from arXiv: 2604.07680 by Cheng Shen, Chenyang Zhang, Jinhong Yuan.

Figure 1
Figure 1. Figure 1: An example for channel matrices in time (left), frequ [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Error performance for different equalization schem [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
read the original abstract

Affine frequency division multiplexing (AFDM) has emerged as a promising waveform for high-mobility communications. However, its equalization remains a practical challenge under general physical channels with off-grid delay and Doppler effects. In this paper, we investigate frequency domain equalization for AFDM by considering a practical filtered-AFDM waveform. We analyze the input-output relations of filtered-AFDM across various domains and show that off-grid effects lead to severe inter-symbol interference in the DAFT domain, limiting the effectiveness of DAFT domain equalization. Motivated by the compactness of the frequency domain channel matrix in wideband systems, we propose a low-complexity two-stage frequency domain equalization scheme. Numerical results demonstrate that the proposed approach achieves performance close to full-block LMMSE equalization with significantly reduced computational complexity, and offers clear advantages over time domain equalization in wideband scenarios.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript analyzes input-output relations for filtered-AFDM under general physical channels with off-grid delay and Doppler, showing that these effects cause severe ISI in the DAFT domain that limits DAFT-domain equalization. Motivated by an observed compactness property of the frequency-domain channel matrix in wideband systems, it proposes a low-complexity two-stage frequency-domain equalization scheme. Numerical results are cited to claim performance close to full-block LMMSE equalization at significantly lower complexity, with advantages over time-domain equalization in wideband scenarios.

Significance. If the frequency-domain compactness holds with negligible approximation error under off-grid effects, the two-stage scheme would provide a practical low-complexity equalizer for high-mobility AFDM communications, addressing a key implementation challenge. The cross-domain input-output analysis is a useful contribution. However, the lack of bounds on leakage or explicit simulation details limits the result's immediate applicability and verifiability.

major comments (2)
  1. [Abstract] Abstract and the section on frequency-domain channel matrix: the central motivation and performance claim rest on the frequency-domain channel matrix remaining 'sufficiently compact' after off-grid delay/Doppler effects, yet no bound on effective bandwidth, leakage energy, or conditions (e.g., maximum delay/Doppler spread relative to bandwidth) is derived to guarantee that two-stage truncation approximates full-block LMMSE with negligible error.
  2. [Numerical results] The section presenting numerical results: the claim that the proposed approach 'achieves performance close to full-block LMMSE equalization with significantly reduced computational complexity' is unsupported by any reported simulation parameters, channel models, error bars, matrix sparsity metrics, or exact complexity counts (e.g., flop counts or big-O expressions), preventing verification of the near-LMMSE claim or the wideband advantage.
minor comments (1)
  1. [Abstract] The abstract would be strengthened by a single sentence quantifying the compactness (e.g., effective bandwidth after leakage) or listing the key simulation parameters used to support the performance claims.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the thorough review and valuable comments on our manuscript. We address each major comment point by point below, indicating where revisions will be made to strengthen the paper.

read point-by-point responses

  1. Referee: [Abstract] Abstract and the section on frequency-domain channel matrix: the central motivation and performance claim rest on the frequency-domain channel matrix remaining 'sufficiently compact' after off-grid delay/Doppler effects, yet no bound on effective bandwidth, leakage energy, or conditions (e.g., maximum delay/Doppler spread relative to bandwidth) is derived to guarantee that two-stage truncation approximates full-block LMMSE with negligible error.

    Authors: We acknowledge the value of explicit bounds for guaranteeing the approximation quality. Our cross-domain analysis demonstrates that off-grid delay and Doppler effects induce severe ISI in the DAFT domain while the frequency-domain channel matrix exhibits compactness in wideband regimes due to the underlying physical channel structure. The two-stage scheme is motivated by this property and supported by numerical evidence of near-LMMSE performance. We will revise the manuscript to add a discussion section clarifying the conditions (relating maximum delay/Doppler spread to system bandwidth) under which compactness is observed. However, deriving rigorous analytical bounds on leakage energy and truncation error under fully general off-grid effects would require substantial additional theoretical work beyond the paper's scope. revision: partial

  2. Referee: [Numerical results] The section presenting numerical results: the claim that the proposed approach 'achieves performance close to full-block LMMSE equalization with significantly reduced computational complexity' is unsupported by any reported simulation parameters, channel models, error bars, matrix sparsity metrics, or exact complexity counts (e.g., flop counts or big-O expressions), preventing verification of the near-LMMSE claim or the wideband advantage.

    Authors: We agree that the current numerical results section lacks sufficient detail for independent verification. In the revised manuscript we will expand this section to report all simulation parameters, the exact channel models (including specific delay and Doppler spread values), error bars from Monte Carlo runs where applicable, quantitative metrics on the effective bandwidth and sparsity of the frequency-domain channel matrix, and precise complexity expressions (including flop counts and big-O notation) comparing the two-stage equalizer to full-block LMMSE and time-domain alternatives. These additions will directly substantiate the performance and complexity claims. revision: yes

standing simulated objections not resolved
  • Deriving rigorous analytical bounds on leakage energy and the approximation error of two-stage truncation under arbitrary off-grid delay/Doppler effects.

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained

full rationale

The paper derives input-output relations for filtered-AFDM directly from the waveform definition and physical channel model, shows DAFT-domain ISI from off-grid effects via this analysis, and motivates the two-stage frequency-domain equalizer from the observed compactness property of the frequency-domain matrix in wideband regimes. Performance closeness to full-block LMMSE is established numerically rather than as a forced prediction. No equation reduces a claimed result to a fitted input by construction, no load-bearing self-citation chain exists in the provided derivation, and the central claims remain independent of the inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim depends on standard wireless-channel modeling assumptions plus the key domain assumption that the frequency-domain matrix stays compact under off-grid effects; no free parameters or new physical entities are introduced in the abstract.

axioms (1)
  • domain assumption The frequency domain channel matrix is compact in wideband systems even with off-grid delay and Doppler shifts
    Explicitly invoked to motivate the two-stage frequency-domain approach.

pith-pipeline@v0.9.0 · 5444 in / 1300 out tokens · 45325 ms · 2026-05-10T18:24:54.288024+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

13 extracted references · 13 canonical work pages · 1 internal anchor

  1. [1]

    Orthogonal time frequency space modulation,

    R. Hadani et al., “Orthogonal time frequency space modulation,” in Proc. of IEEE WCNC , 2017

    work page 2017

  2. [2]

    Orthogonal delay-Doppler division m ultiplexing modulation,

    H. Lin and J. Y uan, “Orthogonal delay-Doppler division m ultiplexing modulation,” IEEE Trans. Wireless Commun., vol. 21, no. 12, pp. 11 024– 11 037, Dec. 2022

    work page 2022

  3. [3]

    Affine frequenc y division multiplexing for next generation wireless communications ,

    A. Bemani, N. Ksairi, and M. Kountouris, “Affine frequenc y division multiplexing for next generation wireless communications ,” IEEE Trans. Wireless Commun., vol. 22, no. 11, pp. 8214–8229, 2023

    work page 2023

  4. [4]

    Low complexity equalization for afdm in doubly disp ersive channels,

    ——, “Low complexity equalization for afdm in doubly disp ersive channels,” in ICASSP, 2022, pp. 5273–5277

    work page 2022

  5. [5]

    Multi-Carrier Modulation: An Evolution from Time-Frequency Domain to Delay-Doppler Domain

    L. Wu et al. , “A message passing detection based affine frequency division multiplexing communication system,” 2023. [Onli ne]. Available: https://doi.org/10.48550/arXiv.2308.01802

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2308.01802 2023

  6. [6]

    Time-domain zero-padding (TZP) AFDM with two-stage iterative mmse detection,

    C. Shen, J. Y uan, and J. Tong, “Time-domain zero-padding (TZP) AFDM with two-stage iterative mmse detection,” IEEE Trans. Wireless Commun., vol. 25, pp. 6255–6269, 2026

    work page 2026

  7. [7]

    Orthogonal delay-doppler division multiplexing (ODDM) over general physical channels,

    J. Tong et al., “Orthogonal delay-doppler division multiplexing (ODDM) over general physical channels,” IEEE Trans. Commun. , pp. 1–1, 2024

    work page 2024

  8. [8]

    Channel-dependent adapt ive time/frequency domain detection for ODDM,

    C. Shen, A. Shafie, and J. Y uan, “Channel-dependent adapt ive time/frequency domain detection for ODDM,” in Proc. IEEE Int. Conf. Commun. (ICC) , 2025, pp. 983–988

    work page 2025

  9. [9]

    Adaptive transmission with frequency-domain precoding and linear equalization over fast fading channels,

    H. Zhang et al., “Adaptive transmission with frequency-domain precoding and linear equalization over fast fading channels,” IEEE Trans. Wireless Commun., vol. 20, no. 11, pp. 7420–7430, 2021

    work page 2021

  10. [10]

    Filtered-OFDM - enabler for flexible waveform in the 5th generation cellular networks,

    X. Zhang et al. , “Filtered-OFDM - enabler for flexible waveform in the 5th generation cellular networks,” in IEEE Globecom , 2015, pp. 1–6

    work page 2015

  11. [11]

    Characterization of randomly time-variant linear channels,

    P . Bello, “Characterization of randomly time-variant linear channels,” IEEE Trans. Commun. Syst. , vol. 11, no. 4, pp. 360–393, 1963

    work page 1963

  12. [12]

    G. H. Golub and C. F. V an Loan, Matrix Computations. Baltimore, MD, USA: Johns Hopkins University Press, 2012, vol. 3

    work page 2012

  13. [13]

    Cross domain iterative detection for orthogonal time frequency space modulation,

    S. Li et al. , “Cross domain iterative detection for orthogonal time frequency space modulation,” IEEE Trans. Wireless Commun. , vol. 21, no. 4, pp. 2227–2242, 2022

    work page 2022