pith. sign in

arxiv: 2604.14666 · v1 · submitted 2026-04-16 · 📡 eess.SP

Low-Complexity Soft-Feedback Detector for AFDM Systems

Taohe Chen , Yin Xu , Tianyao Ma , Aimin Tang , Qu Luo , Dazhi He , Wenjun Zhang This is my paper

Pith reviewed 2026-05-10 11:04 UTC · model grok-4.3

classification 📡 eess.SP
keywords AFDMsoft-feedback detectorMRCdoubly dispersive channelslow-complexity detectionbit error rateiterative detection
0
0 comments X

The pith

A soft-feedback detector for AFDM systems reduces error propagation by using log-likelihood ratio derived soft decisions in an MRC framework.

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

The paper proposes a low-complexity soft-feedback detector for affine frequency division multiplexing in channels with both time and frequency dispersion. It starts from a maximum ratio combining estimator and adds iterative soft feedback from symbol probabilities to improve accuracy without much added computation. This matters because existing detectors either sacrifice performance for speed or become too complex for practical use in mobile high-speed scenarios. If effective, it allows AFDM to deliver its promised full diversity with feasible receivers. The key result is a consistent 3 dB gain over plain MRC at a bit error rate of 10 to the minus 3.

Core claim

The soft-feedback detector incorporates extrinsic information derived from the log-likelihood ratios of transmitted symbols as soft-decision feedback into the MRC estimator, thereby mitigating error propagation during iterative detection and achieving higher detection accuracy at low computational complexity.

What carries the argument

The soft-feedback detector (SFD), which extends the maximum ratio combining estimator by feeding back a priori symbol distributions from log-likelihood ratios to reduce error propagation in iterative detection.

If this is right

  • The detector maintains low complexity while outperforming conventional decision-feedback methods.
  • It enables more reliable data recovery in doubly dispersive channels without perfect channel knowledge assumptions.
  • Simulation results show approximately 3 dB SNR improvement at BER of 10^{-3} compared to MRC.
  • Consistent gains across benchmark detectors suggest broader applicability in AFDM deployments.

Where Pith is reading between the lines

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

  • Such feedback mechanisms could be adapted to other multi-carrier schemes facing similar dispersion issues.
  • Reducing error propagation this way might lower the required transmit power in high-mobility communication systems.
  • Further integration with channel estimation techniques could yield even better performance in real-world imperfect CSI conditions.

Load-bearing premise

The approach assumes that soft-decision feedback from log-likelihood ratios can be added without introducing significant new error propagation in doubly dispersive channels.

What would settle it

Running the detector in a simulation or real test with high Doppler spread and imperfect channel estimates, checking whether the 3 dB gain at 10^{-3} BER still holds or disappears.

Figures

Figures reproduced from arXiv: 2604.14666 by Aimin Tang, Dazhi He, Qu Luo, Taohe Chen, Tianyao Ma, Wenjun Zhang, Yin Xu.

Figure 1
Figure 1. Figure 1: Illustration of the approximate band matrix structure. Gray cells [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The iteration structure of SFD. we proceed to derive the a priori information for the symbol estimation in each iteration to reduce the error propagation, which leverages the a priori distribution of transmitted sym￾bols during iterative feedback. Without loss of generality, the QPSK constellation is adopted to present the design. However, the proposed mechanism can be readily extended to higher-order modu… view at source ↗
Figure 3
Figure 3. Figure 3: Convergence performance comparison. 0 2 4 6 8 10 12 14 16 18 SNR (dB) 10 −4 10 −3 10 −2 10 −1 10 0 BER MMSE MRC-DFE MF-MP SFD [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: BER performance comparison of different detectors. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
read the original abstract

Affine frequency division multiplexing (AFDM), an emerging multi-carrier modulation scheme, has garnered significant attention due to its resilience to Doppler shifts and capability to achieve full diversity in doubly dispersive channels. However, existing data detection algorithms for AFDM systems face a significant trade-off between computational complexity and accuracy. In this paper, a novel low-complexity data detection scheme, termed the soft-feedback detector (SFD), is proposed. Particularly, building upon a maximum ratio combining (MRC) estimator framework, the SFD leverages the a priori symbol distribution to mitigate error propagation during iterative detection. Specifically, soft-decision feedback is incorporated as extrinsic information derived from the log-likelihood ratios of the transmitted symbols. As a result, the proposed detector significantly enhances detection accuracy while maintaining low computational complexity. Simulation results demonstrate that the SFD consistently outperforms benchmark decision-feedback detectors. In particular, compared with the conventional MRC detector, the proposed scheme achieves approximately a 3 dB signal-to-noise ratio (SNR) gain at the bit error rate (BER) of $10^{-3}$.

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 / 2 minor

Summary. The paper proposes a soft-feedback detector (SFD) for affine frequency division multiplexing (AFDM) in doubly dispersive channels. It extends a maximum-ratio combining (MRC) framework by iteratively feeding back extrinsic soft symbols obtained from log-likelihood ratios (LLRs) to reduce error propagation. The central performance claim is that the SFD achieves an approximately 3 dB SNR gain at BER = 10^{-3} relative to conventional MRC while preserving low computational complexity.

Significance. If the reported gain is shown to be robust, the SFD would supply a practical, low-complexity alternative for AFDM detection, exploiting the scheme's full-diversity property in high-mobility settings. The construction re-uses standard soft-information feedback on top of MRC, so its value lies in the specific application and the empirical improvement rather than in a new theoretical framework.

major comments (2)
  1. [Abstract and Simulation Results] Abstract and Simulation Results section: the 3 dB SNR gain at BER = 10^{-3} is the load-bearing performance claim, yet no channel model parameters (Doppler spread relative to subcarrier spacing, power-delay profile), iteration count, exact LLR-to-soft-symbol mapping, or Monte-Carlo statistics (number of trials, error bars) are supplied. Without these, it is impossible to determine whether the gain survives changes in Doppler regime or is an artifact of the particular simulation conditions.
  2. [Detector construction] Detector construction (likely §3): the transition from hard to soft feedback is described, but no convergence analysis, fixed-point characterization, or bound on residual error propagation is given. In high-Doppler regimes where the initial MRC estimates contain substantial ISI, the soft symbols may remain correlated with those errors; the manuscript therefore provides no analytical support for the assertion that soft feedback “mitigates error propagation.”
minor comments (2)
  1. [Simulation Results] Ensure that all simulation curves are accompanied by the exact parameter set (carrier frequency, subcarrier spacing, number of subcarriers, modulation order) so that the experiments are reproducible.
  2. [Notation] Clarify the notation for the soft symbol estimates and the extrinsic LLRs; inconsistent use of hats or tildes can obscure the distinction between a priori and a posteriori quantities.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments, which help improve the clarity and completeness of our work. We address each major comment below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: [Abstract and Simulation Results] Abstract and Simulation Results section: the 3 dB SNR gain at BER = 10^{-3} is the load-bearing performance claim, yet no channel model parameters (Doppler spread relative to subcarrier spacing, power-delay profile), iteration count, exact LLR-to-soft-symbol mapping, or Monte-Carlo statistics (number of trials, error bars) are supplied. Without these, it is impossible to determine whether the gain survives changes in Doppler regime or is an artifact of the particular simulation conditions.

    Authors: We agree that these simulation parameters are essential for reproducibility and for evaluating the robustness of the reported performance gain. In the revised manuscript, we will augment the Simulation Results section with the normalized Doppler spread relative to subcarrier spacing, the power-delay profile, the number of soft-feedback iterations, the exact LLR-to-soft-symbol mapping formula, and the Monte-Carlo trial count together with any error bars or confidence intervals. These parameters follow standard settings for doubly dispersive channel simulations and will be documented explicitly to confirm that the approximately 3 dB gain at BER = 10^{-3} is observed under the stated conditions. revision: yes

  2. Referee: [Detector construction] Detector construction (likely §3): the transition from hard to soft feedback is described, but no convergence analysis, fixed-point characterization, or bound on residual error propagation is given. In high-Doppler regimes where the initial MRC estimates contain substantial ISI, the soft symbols may remain correlated with those errors; the manuscript therefore provides no analytical support for the assertion that soft feedback “mitigates error propagation.”

    Authors: The SFD is an algorithmic extension of the MRC framework that replaces hard decisions with extrinsic soft symbols derived from LLRs to reduce the weight of unreliable feedback and thereby limit error propagation. The current manuscript focuses on the low-complexity implementation and its empirical performance rather than a full theoretical characterization. We acknowledge that a convergence analysis or bound on residual error propagation would provide stronger analytical support, particularly in high-Doppler regimes. Deriving such bounds for time-varying channels is non-trivial and lies outside the scope of this work, which prioritizes practical detection. In the revision we will add a short discussion of the design rationale for using extrinsic information and note that a formal convergence study remains an open direction for future research. The consistent BER gains observed across the simulated scenarios provide practical evidence that soft feedback mitigates error propagation relative to hard-decision MRC. revision: partial

Circularity Check

0 steps flagged

No significant circularity in derivation or performance claim

full rationale

The paper describes a standard iterative detector construction: MRC combiner augmented with extrinsic soft symbols obtained from LLRs. The 3 dB gain at BER=10^{-3} is asserted solely from Monte-Carlo simulations under specific channel models; no equation is shown to reduce the gain to a fitted parameter, a self-referential definition, or a self-citation chain. The algorithm steps (soft feedback incorporation, complexity analysis) are presented as explicit design choices with independent motivation from existing MRC literature. No uniqueness theorem, ansatz smuggling, or renaming of known results is invoked to force the result. The derivation chain therefore remains self-contained against external benchmarks and does not collapse by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on standard domain assumptions about doubly dispersive channels and the utility of soft information; no new free parameters or invented entities are introduced in the abstract.

axioms (1)
  • domain assumption The underlying channel is doubly dispersive and the receiver has access to channel estimates sufficient for MRC.
    Invoked to justify the resilience of AFDM and the applicability of the MRC estimator framework.

pith-pipeline@v0.9.0 · 5498 in / 1188 out tokens · 54631 ms · 2026-05-10T11:04:57.448406+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

8 extracted references · 8 canonical work pages

  1. [1]

    ITU-R framework for IMT-2030: Review and future direction,

    NGMN Alliance, “ITU-R framework for IMT-2030: Review and future direction,” NGMN Alliance, Düsseldorf, Germany, Tech. Rep., Feb. 2024

    work page 2030

  2. [2]

    Affine frequency division multiplexing for next generation wireless communications,

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

    work page 2023

  3. [3]

    Low complexity equalization for AFDM in doubly dispersive channels,

    ——, “Low complexity equalization for AFDM in doubly dispersive channels,” inProc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP), 2022, pp. 5273–5277

    work page 2022

  4. [4]

    AFDM signal detection based on message passing scheme,

    L. Wu, S. Luo, D. Song, F. Yang, R. Lin, and S. Xie, “AFDM signal detection based on message passing scheme,”Digit. Signal Process., vol. 153, p. 104633, 2024

    work page 2024

  5. [5]

    Turbo equalization scheme for Polar-AFDM system,

    X. Qu, X. Shen, H. Wang, and X. Li, “Turbo equalization scheme for Polar-AFDM system,” inProc. Int. Conf. Commun. Softw. Netw. (ICCSN), 2024, pp. 278–282

    work page 2024

  6. [6]

    Minimum mean squared error equalization using a priori information,

    M. Tuchler, A. C. Singer, and R. Koetter, “Minimum mean squared error equalization using a priori information,”IEEE Trans. Signal Process., vol. 50, no. 3, pp. 673–683, 2002

    work page 2002

  7. [7]

    Low-complexity block turbo equaliza- tion for OFDM systems in time-varying channels,

    K. Fang, L. Rugini, and G. Leus, “Low-complexity block turbo equaliza- tion for OFDM systems in time-varying channels,”IEEE Trans. Signal Process., vol. 56, no. 11, pp. 5555–5566, 2008

    work page 2008

  8. [8]

    A twofold lookup table architecture for efficient approximation of activation functions,

    Y . Xie, A. N. J. Raj, Z. Hu, S. Huang, Z. Fan, and M. Joler, “A twofold lookup table architecture for efficient approximation of activation functions,”IEEE Trans. VLSI Syst., vol. 28, no. 12, pp. 2540–2550, 2020

    work page 2020