arxiv: 2605.00049 · v1 · submitted 2026-04-29 · 💻 cs.IT · math.IT

Recognition: unknown

Delay-Doppler Domain Channel Estimation: What if Sparsity is Unknown?

Zijian Yang , Yulin Shao , Fen Hou , Shaodan Ma

Authors on Pith no claims yet

Pith reviewed 2026-05-09 20:29 UTC · model grok-4.3

classification 💻 cs.IT math.IT

keywords delay-Doppler channel estimationsparsity-agnostic estimationCartesian-product supportBayesian information criterionaffine frequency division multiplexinghigh-mobility communicationsstructured sparsity

0 comments

The pith

A sparsity-agnostic estimator recovers exact delay-Doppler channel support by selecting dimensions via the Bayesian information criterion.

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

The paper establishes a channel estimator for the delay-Doppler domain that needs no advance knowledge of how many delays or Dopplers are active. It works by assuming active delays share one common set of Dopplers, forming a Cartesian-product support, then letting the Bayesian information criterion pick the right number of each from the received data. This is shown on an affine frequency division multiplexing system whose observation model fits an on-grid delay-Doppler representation. A reader should care because real channels have time-varying and unknown sparsity, so fixed-budget or oracle-knowing methods often fail or waste resources, while this method adapts automatically.

Core claim

By exploiting the Cartesian-product structure of the delay-Doppler support and selecting its dimensions directly from the data with the Bayesian information criterion, the estimator recovers the exact support with high probability and achieves near-oracle channel reconstruction accuracy without any prior sparsity budgets.

What carries the argument

Sparsity-agnostic structured estimator that uses the Cartesian-product structure of DD support together with Bayesian information criterion selection of delay and Doppler dimensions.

If this is right

Exact support recovery occurs with high probability.
Channel reconstruction accuracy approaches that of an oracle that knows the true sparsity.
Performance stays above fixed-budget methods and sparse Bayesian learning across tested conditions.
The estimator works for any waveform whose model admits the required on-grid representation.
It provides an adaptive solution when sparsity changes over time.

Where Pith is reading between the lines

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

The same dimension-selection idea could be tested on other multicarrier waveforms if an on-grid delay-Doppler mapping can be derived.
In high-mobility links the method might lower the required pilot density because it no longer needs conservative upper bounds on sparsity.
The Cartesian-product assumption could be relaxed in future work by allowing multiple Doppler sets per delay cluster while still using information criteria for selection.

Load-bearing premise

The active delays must share exactly one common set of Dopplers and the received signal must fit an on-grid delay-Doppler model.

What would settle it

Run the estimator on AFDM data with randomly varying sparsity levels that obey the Cartesian-product structure and check whether the probability of exact support recovery falls significantly below the reported high value.

Figures

Figures reproduced from arXiv: 2605.00049 by Fen Hou, Shaodan Ma, Yulin Shao, Zijian Yang.

**Figure 3.** Figure 3: Comparison of the true and recovered DD supports for one representative realization at [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗

read the original abstract

Sparsity in the delay-Doppler (DD) domain enables efficient channel estimation, but the realization-wise sparsity level is rarely known in advance, and it fluctuates. What if we could estimate the channel without ever knowing how many delays or Dopplers are active? This paper answers that question. We propose a sparsity-agnostic structured estimator that requires no prior knowledge of delay or Doppler sparsity budgets. The key idea is to exploit the Cartesian-product structure of DD support (active delays share a common Doppler set) and to select the support dimensions directly from the data via the Bayesian information criterion. We instantiate the framework on an affine frequency division multiplexing system, where the observation model naturally admits an on-grid DD representation. Numerical results demonstrate that it recovers the exact support with high probability and achieves near-oracle channel reconstruction accuracy, consistently outperforming fixed-budget baselines and sparse Bayesian learning. The approach is waveform-agnostic and offers a practical, adaptive solution for DD-domain channel estimation under unknown and time-varying sparsity.

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.

Desk Editor's Note private letter to a colleague

The paper's main move is a BIC selector for the number of active delays and Dopplers in DD estimation, built on the assumption that the support is exactly a Cartesian product.

read the letter

The useful piece is the data-driven choice of support dimensions via BIC after restricting the search to Cartesian-product sets. That removes the need to guess the sparsity budgets ahead of time, and the AFDM simulations show the method recovering the exact support most of the time and getting close to oracle NMSE while beating fixed-budget estimators and SBL. The structure exploitation shrinks the search space enough to make the BIC step practical, which is a concrete engineering step forward for receivers that see time-varying sparsity.

Referee Report

1 major / 2 minor

Summary. The manuscript proposes a sparsity-agnostic structured estimator for delay-Doppler (DD) domain channel estimation that exploits the Cartesian-product structure of DD support (active delays share a common Doppler set) and selects the support dimensions from the data via the Bayesian information criterion (BIC), without prior knowledge of sparsity levels. The framework is instantiated on an affine frequency division multiplexing (AFDM) system admitting an on-grid DD representation. Numerical results claim that the estimator recovers the exact support with high probability, achieves near-oracle channel reconstruction accuracy, and outperforms fixed-budget baselines and sparse Bayesian learning.

Significance. If the results hold under the stated modeling assumptions, the work offers a practical adaptive solution for DD-domain channel estimation when sparsity is unknown and time-varying, which is a common practical challenge. The waveform-agnostic framing and direct data-driven dimension selection via BIC are positive features that could extend to other systems if the Cartesian-product structure is appropriate.

major comments (1)

[Estimator design and numerical results sections] The performance claims (exact support recovery with high probability and near-oracle NMSE) are load-bearing on the Cartesian-product support assumption, which is invoked both to reduce the search space for BIC selection and to derive the reported accuracy. In general DD channels each path can have an independent Doppler shift, yielding an arbitrary sparse support rather than a product set; the paper should therefore state the precise conditions under which the structure holds and include a robustness experiment (or counter-example) when the assumption is violated, as this directly affects whether the BIC criterion remains correctly specified and the numerical evidence generalizes.

minor comments (2)

[Numerical results] The abstract and results sections report positive numerical outcomes but omit full details on simulation setups, error bars, exact baseline implementations, and derivation steps for the AFDM observation model; adding these would strengthen verifiability without altering the central claims.
[Throughout] Notation for the on-grid DD representation and the precise form of the BIC objective should be defined explicitly at first use to avoid ambiguity for readers unfamiliar with the AFDM instantiation.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive and insightful comments on our manuscript. We address the major comment below and will revise the manuscript to incorporate the suggested clarifications and additional analysis.

read point-by-point responses

Referee: [Estimator design and numerical results sections] The performance claims (exact support recovery with high probability and near-oracle NMSE) are load-bearing on the Cartesian-product support assumption, which is invoked both to reduce the search space for BIC selection and to derive the reported accuracy. In general DD channels each path can have an independent Doppler shift, yielding an arbitrary sparse support rather than a product set; the paper should therefore state the precise conditions under which the structure holds and include a robustness experiment (or counter-example) when the assumption is violated, as this directly affects whether the BIC criterion remains correctly specified and the numerical evidence generalizes.

Authors: We agree that the Cartesian-product support assumption is fundamental to both the estimator design (reducing the BIC search space) and the performance claims. In the revised manuscript, we will add an explicit statement in the introduction and estimator design sections clarifying the precise conditions under which the structure holds: namely, in DD-domain channels where the set of active delays shares a common Doppler support set, which arises naturally in the AFDM on-grid representation under the assumed multipath model with correlated scattering across delays. We will also include a new subsection in the numerical results with a robustness experiment that perturbs the assumption by introducing independent Doppler shifts for a fraction of paths, along with a counter-example demonstrating performance degradation and the resulting mismatch in the BIC criterion. This will delineate the scope of the claims and the conditions for the reported high-probability exact support recovery. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained under explicit modeling assumptions

full rationale

The paper's core procedure selects support dimensions via BIC applied to data under the explicitly stated Cartesian-product structure of DD support. This is a standard model-selection step whose output is not equivalent to its inputs by construction, nor is any performance metric (exact support recovery probability or NMSE) redefined as a fitted quantity. Numerical results are presented as empirical evidence conditional on the on-grid observation model and product structure; they do not reduce to tautology. No self-citation chain, ansatz smuggling, or renaming of known results is load-bearing in the derivation. The approach remains falsifiable outside the fitted values and is therefore scored as non-circular.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Based on abstract only: the central claim rests on the domain assumption of Cartesian-product DD support structure and on-grid representation in the chosen waveform.

axioms (2)

domain assumption DD support exhibits Cartesian-product structure where active delays share a common Doppler set
Invoked as the key structural property enabling the estimator.
domain assumption Observation model admits on-grid DD representation
Stated for the AFDM system instantiation.

pith-pipeline@v0.9.0 · 5478 in / 1298 out tokens · 25144 ms · 2026-05-09T20:29:27.239700+00:00 · methodology

discussion (0)

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Stepped Frequency Division Multiplexing: A Jump-Free Continuous-Time AFDM Waveform
eess.SP 2026-05 unverdicted novelty 6.0

SFDM creates a continuous AFDM waveform by holding instantaneous frequency at the wrapped chirp midpoint per interval while accumulating phase continuously, eliminating jumps, lowering out-of-band emissions, and prese...

Reference graph

Works this paper leans on

15 extracted references · 1 canonical work pages · cited by 1 Pith paper

[1]

Compressed channel sensing: A new approach to estimating sparse multipath chan- nels,

W. U. Bajwa, J. Haupt, A. M. Sayeed, and R. Nowak, “Compressed channel sensing: A new approach to estimating sparse multipath chan- nels,”Proceedings of the IEEE, vol. 98, no. 6, pp. 1058–1076, 2010

2010
[2]

Geometry-based directional model for mobile radio chan- nels—principles and implementation,

A. F. Molisch, A. Kuchar, J. Laurila, K. Hugl, and R. Schmalen- berger, “Geometry-based directional model for mobile radio chan- nels—principles and implementation,”European Transactions on Telecommunications, vol. 14, no. 4, pp. 351–359, 2003

2003
[3]

Polarization-aware movable antenna,

R. Zhang, Y . Shao, and Y . C. Eldar, “Polarization-aware movable antenna,”IEEE Transactions on Wireless Communications, vol. 25, pp. 7428–7442, 2025

2025
[4]

Compressive estimation of doubly selective channels in multicarrier systems: Leakage effects and sparsity-enhancing processing,

G. Taubock, F. Hlawatsch, D. Eiwen, and H. Rauhut, “Compressive estimation of doubly selective channels in multicarrier systems: Leakage effects and sparsity-enhancing processing,”IEEE Journal of Selected Topics in Signal Processing, vol. 4, no. 2, pp. 255–271, 2010

2010
[5]

Application of com- pressive sensing to sparse channel estimation,

C. R. Berger, Z. Wang, J. Huang, and S. Zhou, “Application of com- pressive sensing to sparse channel estimation,”IEEE Communications Magazine, vol. 48, no. 11, pp. 164–174, 2010

2010
[6]

Y . C. Eldar and G. Kutyniok,Compressed sensing: theory and applica- tions. Cambridge university press, 2012

2012
[7]

Signal recovery from random mea- surements via orthogonal matching pursuit,

J. A. Tropp and A. C. Gilbert, “Signal recovery from random mea- surements via orthogonal matching pursuit,”IEEE Transactions on information theory, vol. 53, no. 12, pp. 4655–4666, 2007

2007
[8]

Models, methods, and waveforms for estimation and prediction of sparse time–varying channels,

W. Benzine, A. Bemani, N. Ksairi, and D. Slock, “Models, methods, and waveforms for estimation and prediction of sparse time–varying channels,”IEEE Transactions on Wireless Communications, vol. 25, pp. 9623–9638, 2025

2025
[9]

Structured compressed sensing: From theory to applications,

M. F. Duarte and Y . C. Eldar, “Structured compressed sensing: From theory to applications,”IEEE Transactions on Signal Processing, vol. 59, no. 9, pp. 4053–4085, 2011

2011
[10]

Study on Channel Model for Frequencies from 0.5 to 100 GHz,

3GPP, “Study on Channel Model for Frequencies from 0.5 to 100 GHz,” 3rd Generation Partnership Project (3GPP), Technical Report TR 38.901, Apr. 2026, Version 19.3.0, Release 19

2026
[11]

Estimating the dimension of a model,

G. Schwarz, “Estimating the dimension of a model,”The Annals of Statistics, vol. 6, no. 2, pp. 461–464, 1978

1978
[12]

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- actions on Wireless Communications, vol. 22, no. 11, pp. 8214–8229, 2023

2023
[13]

Agile affine frequency division multiplexing,

Y . Cao and Y . Shao, “Agile affine frequency division multiplexing,” arXiv preprint arXiv:2512.14424, 2025

work page arXiv 2025
[14]

Federated edge learning with misaligned over-the-air computation,

Y . Shao, D. G ¨und¨uz, and S. C. Liew, “Federated edge learning with misaligned over-the-air computation,”IEEE Transactions on Wireless Communications, vol. 21, no. 6, pp. 3951–3964, 2021

2021
[15]

Ridge regression: Biased estimation for nonorthogonal problems,

A. E. Hoerl and R. W. Kennard, “Ridge regression: Biased estimation for nonorthogonal problems,”Technometrics, vol. 12, no. 1, pp. 55–67, 1970

1970