Pith. sign in

REVIEW 1 major objections 48 references

A Brownian bridge diffusion model jointly estimates channels and detects data after suppressing jamming in the time-frequency domain.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · grok-4.3

2026-06-30 08:53 UTC pith:3B23SCUX

load-bearing objection The paper pairs STFT jamming suppression with a Brownian bridge diffusion stage for joint estimation and detection, but the modeling choice lacks separate validation. the 1 major comments →

arxiv 2606.28778 v1 pith:3B23SCUX submitted 2026-06-27 cs.IT cs.AIeess.SPmath.IT

Brownian Bridge Diffusion-Based Joint Channel Estimation and Data Detection for Jamming-Resilient Receivers

Honghan She , Yufan Cheng , Tieming Sun , Pengyu Wang , Siya Huang , Kaikai Yang This is my paper
classification cs.IT cs.AIeess.SPmath.IT
keywords jammingchannel estimationdata detectionBrownian bridge diffusionSTFTjoint estimationwireless receiver
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper proposes a framework that first extracts jamming features in the short-time Fourier transform domain and suppresses those samples to raise the signal-to-jamming-plus-noise ratio. It then applies a Brownian bridge diffusion process to model the evolution of the cleaned received signal together with the encoded bits, even when channel estimates contain errors, so that channel estimation and data detection can be performed jointly. A fast ordinary differential equation solver is derived to keep the iterative evolution low in complexity, and a multi-module training procedure is introduced to strengthen recovery. The approach targets the case in which jamming overlaps pilots and data in both time and frequency, a situation that defeats many existing receivers. Simulations indicate that the resulting receiver recovers bits more accurately than baseline methods while using fewer parameters and comparable computation.

Core claim

The Brownian bridge diffusion-based joint channel estimation and data detection framework extracts jamming features in the STFT domain to suppress interference and then applies a BBD process to model the evolution of the jamming-suppressed signal and encoded bits, enabling enhanced joint estimation and detection with a derived fast ODE solver.

What carries the argument

The Brownian bridge diffusion (BBD) process, which models the evolution of the jamming-suppressed signal and the encoded bits in the presence of channel estimation errors.

Load-bearing premise

The Brownian bridge diffusion process can accurately model the evolution of the jamming-suppressed signal and the encoded bits in the presence of channel estimation errors.

What would settle it

A simulation or hardware experiment in which the proposed receiver fails to produce lower bit error rates than the listed baselines under overlapping time-frequency jamming would falsify the performance claim.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • The framework achieves superior bit recovery performance compared with baseline schemes.
  • It maintains a lower number of model parameters than the baselines.
  • It exhibits competitive computational complexity.
  • The derived fast ODE solver reduces the complexity of the BBD iterative evolution.

Where Pith is reading between the lines

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

  • The same two-stage structure of STFT suppression followed by diffusion modeling could be tested on other structured interference such as narrowband or pulsed jamming.
  • Replacing the fixed multi-module training with an online variant might allow the receiver to track time-varying jamming statistics without retraining.
  • The BBD modeling step could be combined with existing pilot-based estimators to form hybrid receivers that fall back to classical methods when diffusion assumptions weaken.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

1 major / 0 minor

Summary. The paper proposes a Brownian bridge diffusion-based joint channel estimation and data detection (BBD-JCED) framework for jamming-resilient receivers in wireless networks. It consists of an STFT-domain module to extract and suppress jamming features, followed by a Brownian bridge diffusion process to jointly model the evolution of the jamming-suppressed signal and encoded bits under channel estimation errors. A fast ODE solver is derived to reduce complexity, and a multi-module training algorithm is introduced. Simulations are reported to show superior bit recovery performance versus baselines, with fewer model parameters and competitive complexity.

Significance. If the BBD modeling assumption is valid and the simulations are rigorously supported, the work could introduce a diffusion-based paradigm for joint estimation/detection under overlapping jamming, potentially offering parameter efficiency. However, the significance is tempered by the absence of any derivation justifying the bridge SDE choice or ablation isolating modeling fidelity from end-to-end effects.

major comments (1)
  1. [Abstract] Abstract: the central claim that the Brownian bridge diffusion process 'enables enhanced joint channel estimation and data detection' by modeling the suppressed signal and encoded bits rests on an unvalidated modeling assumption. No derivation is supplied showing why the bridge SDE (as opposed to standard diffusion or other processes) correctly captures the dynamics in the presence of residual channel estimation errors, nor is any ablation mentioned that separates modeling accuracy from hyperparameter or training effects. This is load-bearing for the performance claims.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thoughtful review and for highlighting the need for stronger justification of the modeling choices. We address the major comment below and commit to revisions that directly respond to the concerns raised.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that the Brownian bridge diffusion process 'enables enhanced joint channel estimation and data detection' by modeling the suppressed signal and encoded bits rests on an unvalidated modeling assumption. No derivation is supplied showing why the bridge SDE (as opposed to standard diffusion or other processes) correctly captures the dynamics in the presence of residual channel estimation errors, nor is any ablation mentioned that separates modeling accuracy from hyperparameter or training effects. This is load-bearing for the performance claims.

    Authors: We acknowledge that the manuscript does not supply an explicit derivation of the bridge SDE or an ablation isolating its contribution. The Brownian bridge is chosen because its fixed-endpoint conditioning property directly supports joint modeling of the jamming-suppressed observation and the encoded bits under channel uncertainty; this is described at a high level in Section II-B. However, we agree a self-contained derivation from the conditional density and a comparison against standard diffusion are absent. In the revision we will insert a new subsection deriving the bridge SDE and add an ablation study that replaces the bridge with a standard diffusion process while keeping all other modules and training identical. These additions will be placed in the main text and will be supported by new simulation results. revision: yes

Circularity Check

0 steps flagged

No circularity detected; derivation self-contained against external benchmarks

full rationale

The provided abstract and description outline a two-module framework (STFT jamming suppression followed by BBD modeling of suppressed signals and bits) plus an ODE solver and multi-module training, but contain no equations, fitted parameters presented as predictions, self-citations used as load-bearing uniqueness theorems, or ansatzes smuggled via prior work. No step reduces by construction to its inputs; the modeling choice and performance claims are presented as independent contributions validated via simulation against baselines. This is the normal case of a self-contained proposal without the enumerated circular patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

Only abstract available; the Brownian bridge diffusion modeling choice and the assumption that STFT suppression improves SJNR are the main unverified modeling steps. No explicit free parameters or axioms are stated in the provided text.

invented entities (1)
  • Brownian bridge diffusion process for joint channel and data modeling no independent evidence
    purpose: To model evolution of suppressed signal and encoded bits under channel estimation errors
    Introduced as the core of the second module without independent evidence or falsifiable prediction outside the framework itself.

pith-pipeline@v0.9.1-grok · 5818 in / 1178 out tokens · 38827 ms · 2026-06-30T08:53:17.414693+00:00 · methodology

0 comments
read the original abstract

In next-generation wireless networks, the growing density of devices and limited spectrum resources pose severe jamming challenges to fragile legitimate communication links in the wireless electromagnetic environment. Crucially, when jamming overlaps with pilot and data symbols in both time and frequency domains, it inflicts a severe bottleneck on receiver-side joint estimation and detection. Existing schemes often lack an effective framework to combat such jamming contamination, thereby failing to guarantee reliable transmission. To address this issue, we propose a Brownian bridge diffusion-based joint channel estimation and data detection framework (BBD-JCED) for jamming-resilient receivers. Specifically, the proposed framework comprises two core modules: the first extracts jamming features in the short-time Fourier transform (STFT) domain and suppresses jamming samples, thereby improving the signal-to-jamming-plus-noise ratio (SJNR) of the received signal; the second introduces a Brownian bridge diffusion (BBD) process to model the evolution of the suppressed signal and the encoded bits in the presence of channel estimation errors, thereby enabling enhanced joint channel estimation and data detection. To alleviate the computational burden of the BBD process in the second module, we further derive a fast ordinary differential equation (ODE) solver that enables its low-complexity iterative evolution. Finally, we design a multi-module training algorithm to improve the data recovery capability of the proposed framework. Simulation results demonstrate that the proposed framework achieves superior bit recovery performance compared with baseline schemes while maintaining a lower number of model parameters and competitive computational complexity.

Figures

Figures reproduced from arXiv: 2606.28778 by Honghan She, Kaikai Yang, Pengyu Wang, Siya Huang, Tieming Sun, Yufan Cheng.

Figure 1
Figure 1. Figure 1: Wireless jamming scenario. The yellow dashed line represents the [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Exemplary OFDM time-frequency grid and time-domain waveform. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Time-frequency spectrum of CSN jamming and LFM jamming. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Overall architecture of the proposed BBD-JCED. First, a JSF module is utilized to improve the SJNR of the received signal. Next, a BB-JR module [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Structure of the JSF module. A U-Net-based mask estimator is trained [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Structure of the BB-JR module. An RCAN-based CSI interpolator is utilized to refine the LS estimate into the interpolated CSI, and a BRL-based [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparison of the standard diffusion process and the BBD process. [PITH_FULL_IMAGE:figures/full_fig_p006_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: BER comparison under CSN jamming with 40 combs. All baseline schemes are equipped with the JSF module. -50 -45 -40 -35 -30 -25 -20 SJR (dB) 0.1 0.2 0.3 0.4 0.5 TraditionRx [8] (with JSF) DECNN [18] (with JSF) CE-CCRNet [21] (with JSF) DM [30] (with JSF) Proposed BBD-JCED Channel BER (a) Channel BER in TDL-A channel -50 -45 -40 -35 -30 -25 -20 SJR (dB) 10 -6 10 -5 10 -4 10 -3 10 -2 10 -1 Data BER TraditionR… view at source ↗
Figure 9
Figure 9. Figure 9: BER comparison under LFM jamming with 6 periods. All baseline schemes are equipped with the JSF module. superior robustness of the proposed framework for jamming￾resilient wireless reception [PITH_FULL_IMAGE:figures/full_fig_p011_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Ablation result of the proposed BBD-JCED under CSN jamming with [PITH_FULL_IMAGE:figures/full_fig_p012_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Ablation result of the proposed BBD-JCED under LFM jamming with [PITH_FULL_IMAGE:figures/full_fig_p012_11.png] view at source ↗

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

48 extracted references

  1. [1]

    From 5G to 6G networks: A survey on AI-based jamming and interference detection and mitigation,

    P. Lohan, B. Kantarci, M. Amine Ferrag, N. Tihanyi, and Y . Shi, “From 5G to 6G networks: A survey on AI-based jamming and interference detection and mitigation,”IEEE Open J. Commun. Soc., vol. 5, pp. 3920– 3974, Jun. 2024

    2024

  2. [2]

    Jamming attacks and anti-jamming strategies in wireless networks: A comprehensive survey,

    H. Pirayesh and H. Zeng, “Jamming attacks and anti-jamming strategies in wireless networks: A comprehensive survey,”IEEE Commun. Surveys Tuts., vol. 24, no. 2, pp. 767–809, Mar. 2022

    2022

  3. [3]

    Interference sup- pression using deep learning: Current approaches and open challenges,

    T. Oyedare, V . K. Shah, D. J. Jakubisin, and J. H. Reed, “Interference sup- pression using deep learning: Current approaches and open challenges,” IEEE Access, vol. 10, pp. 66238–66266, Jun. 2022

    2022

  4. [4]

    Secure physical layer communications for low-altitude economy networking: A survey,

    L. Caiet al., “Secure physical layer communications for low-altitude economy networking: A survey,”IEEE Commun. Surveys Tuts., vol. 28, pp. 2497-2530, Nov. 2025

    2025

  5. [5]

    A survey on jamming techniques in physical layer security and anti-jamming strategies for 6G,

    K. N. Vaishnavi, S. D. Khorvi, R. Kishore, and S. Gurugopinath, “A survey on jamming techniques in physical layer security and anti-jamming strategies for 6G,” inProc. IEEE Int. Conf. Telecommun. (ICT), London, United Kingdom, Jun. 2021, pp. 174–179

    2021

  6. [6]

    Intelligent wireless interference recognition: Challenges, opportunities, and future directions,

    P. Wang, X. Huang, H. Zhang, and Z. Wang, “Intelligent wireless interference recognition: Challenges, opportunities, and future directions,” IEEE Wireless Commun., vol. 32, no. 6, pp. 209–215, Oct. 2025

    2025

  7. [7]

    Dynamic spectrum anti-jamming communications: Challenges and opportunities,

    X. Wang, J. Wang, Y . Xu, J. Chen, L. Jia, and X. Liu, “Dynamic spectrum anti-jamming communications: Challenges and opportunities,” IEEE Commun. Mag., vol. 58, no. 2, pp. 79–85, Feb. 2020. 13

    2020

  8. [8]

    Detection, estimation, and modulation theory, part I: Detection, estimation, and linear modulation theory,

    H. L. Van Trees, “Detection, estimation, and modulation theory, part I: Detection, estimation, and linear modulation theory,” Hoboken, NJ, USA: Wiley, Apr. 2004

    2004

  9. [9]

    Digital communications, 5th edition,

    J. G. Proakis and M. Salehi, “Digital communications, 5th edition,” Manhattan, New York, USA: McGraw-Hill, Nov. 2007

    2007

  10. [10]

    Intelligent wireless interference iden- tification with lightweight transformer network,

    P. Wang, Z. Wang, and Z. Han, “Intelligent wireless interference iden- tification with lightweight transformer network,”IEEE Trans. Commun., vol. 73, no. 11, pp. 11355–11367, Jul. 2025

    2025

  11. [11]

    Jamming identification with differential transformer for low-altitude wireless networks,

    P. Wang, Z. Wang, T. Mao, W. Yuan, H. Zhang, and G. K. Karagiannidis, “Jamming identification with differential transformer for low-altitude wireless networks,”IEEE J. Sel. Areas Commun., vol. 44, pp. 3664–3677, Mar. 2026

    2026

  12. [12]

    Deep learning-based channel estimation,

    M. Soltani, V . Pourahmadi, A. Mirzaei, and H. Sheikhzadeh, “Deep learning-based channel estimation,”IEEE Commun. Lett., vol. 23, no. 4, pp. 652–655, Feb. 2019

    2019

  13. [13]

    Deep residual learning meets OFDM channel estimation,

    L. Li, H. Chen, H. -H. Chang, and L. Liu, “Deep residual learning meets OFDM channel estimation,”IEEE Wireless Commun. Lett., vol. 9, no. 5, pp. 615–618, Dec. 2019

    2019

  14. [14]

    Attention based neural networks for wireless channel estimation,

    D. Luan and J. Thompson, “Attention based neural networks for wireless channel estimation,” inProc. IEEE V eh. Technol. Conf. (VTC), Helsinki, Finland, Jun. 2022, pp. 1–5

    2022

  15. [15]

    AE-DENet: Enhancement for deep learning-based channel estimation in OFDM systems,

    E. Fola, Y . Luo, and C. Luo, “AE-DENet: Enhancement for deep learning-based channel estimation in OFDM systems,” inProc. IEEE Global Telecommun. Conf. (GLOBECOM), Cape Town, South Africa, Dec. 2024, pp. 1449–1454

    2024

  16. [16]

    AdaFortiTran: An adaptive transformer model for robust OFDM channel estimation,

    B. Guler and H. Jafarkhani, “AdaFortiTran: An adaptive transformer model for robust OFDM channel estimation,” inProc. IEEE Int. Conf. Commun. (ICC), Montreal, QC, Canada, Jun. 2025, pp. 3797–3802

    2025

  17. [17]

    Power of deep learning for channel estimation and signal detection in OFDM systems,

    H. Ye, G. Y . Li, and B. -H. Juang, “Power of deep learning for channel estimation and signal detection in OFDM systems,” inIEEE Wireless Commun. Lett., vol. 7, no. 1, pp. 114–117, Sep. 2017

    2017

  18. [18]

    CNN-based underwater acoustic OFDM communications over doubly-selective channels,

    J. Liu, F. Ji, H. Zhao, J. Li, and M. Wen, “CNN-based underwater acoustic OFDM communications over doubly-selective channels,” in Proc. IEEE V eh. Technol. Conf. (VTC), Norman, OK, USA, Sep. 2021, pp. 01–06

    2021

  19. [19]

    TransDetector: A transformer-based detector for underwater acoustic differential OFDM communications,

    Y . Li, S. Wang, D. Liu, C. Zhou, and Z. Gui, “TransDetector: A transformer-based detector for underwater acoustic differential OFDM communications,”IEEE Trans. Wireless Commun., vol. 23, no. 8, pp. 9899–9911, Feb. 2024

    2024

  20. [20]

    ComNet: Combination of deep learning and expert knowledge in OFDM receivers,

    X. Gao, S. Jin, C. -K. Wen, and G. Y . Li, “ComNet: Combination of deep learning and expert knowledge in OFDM receivers,”IEEE Commun. Lett., vol. 22, no. 12, pp. 2627–2630, Oct. 2018

    2018

  21. [21]

    Deep learning for joint channel estimation and signal detection in OFDM systems,

    X. Yi and C. Zhong, “Deep learning for joint channel estimation and signal detection in OFDM systems,”IEEE Commun. Lett., vol. 24, no. 12, pp. 2780–2784, Aug. 2020

    2020

  22. [22]

    Better late than never: GAN-enhanced dynamic anti- jamming spectrum access with incomplete sensing information,

    H. Hanet al., “Better late than never: GAN-enhanced dynamic anti- jamming spectrum access with incomplete sensing information,”IEEE Wireless Commun. Lett., vol. 10, no. 8, pp. 1800–1804, May. 2021

    2021

  23. [23]

    Joint interference cancellation and signal detection using latent space representations in V AE,

    I. W. C. Wai, M. H. Jaward, V . M. Baskaran, S. -N. Liang, C. C. C. Hin, and M. L. Sim, “Joint interference cancellation and signal detection using latent space representations in V AE,”IEEE Trans. Consum. Electron., vol. 70, no. 1, pp. 197–208, Oct. 2023

    2023

  24. [24]

    Generative diffusion models for wireless networks: Fundamental, architecture, and state-of-the-art,

    D. Fanet al., “Generative diffusion models for wireless networks: Fundamental, architecture, and state-of-the-art,”IEEE Commun. Surveys Tuts., vol. 28, pp. 5632–5677, Mar. 2026

    2026

  25. [25]

    Generative AI for secure physical layer communications: A survey,

    C. Zhaoet al., “Generative AI for secure physical layer communications: A survey,”IEEE Commun. Surveys Tuts., vol. 11, no. 1, pp. 3–26, Aug. 2024

    2024

  26. [26]

    Score-based generative modeling for interference mitigation in automotive FMCW radar,

    X. Wei, J. Overdevest, J. Li, J. Youn, S. Ravindran, and R. J. G. Van Sloun, “Score-based generative modeling for interference mitigation in automotive FMCW radar,” inProc. Eur . Radar Conf. (EuRAD), Paris, France, Sep. 2024, pp. 27–30

    2024

  27. [27]

    Model- based diffusion for mitigating automotive radar interference,

    J. Overdevest, X. Wei, H. van Gorp, and R. J. G. van Sloun, “Model- based diffusion for mitigating automotive radar interference,” inProc. IEEE Int. Conf. Acoust., Speech, Signal Process. Workshop. (ICASSPW), Seoul, Korea, Republic of, Apr. 2024, pp. 284–288

    2024

  28. [28]

    ICDM: Interference cancellation diffusion models for wireless semantic communications,

    T. Wuet al., “ICDM: Interference cancellation diffusion models for wireless semantic communications,”IEEE J. Sel. Areas Commun., vol. 44, pp. 2528–2543, Dec. 2025

    2025

  29. [29]

    RadarImpu- tationNet: An enhanced long-term diffusion model for automotive radar interference suppression and signal recovery,

    M. Yuan, M. Lian, Y . Zhang, C. Li, J. Shi and T. Zhang, “RadarImpu- tationNet: An enhanced long-term diffusion model for automotive radar interference suppression and signal recovery,”IEEE Sensors J., vol. 25, no. 13, pp. 24693–24711, July, 2025

    2025

  30. [30]

    Conditional diffusion models for signal reconstruction after interference notching,

    H. She, Y . Cheng, Y . Tang, and Y . Zhao, “Conditional diffusion models for signal reconstruction after interference notching,”IEEE Wireless Commun. Lett., vol. 15, pp. 2040–2044, Feb. 2026

    2040

  31. [31]

    Denoising diffusion probabilistic models,

    J. Ho, A. Jain, and P. Abbeel, “Denoising diffusion probabilistic models,” inProc. Adv. Neural Inf. Process. Syst. (NeurIPS), Vancouver, British Columbia, Canada (Virtual conference), Dec. 2020, pp. 6840–6851

    2020

  32. [32]

    Denoising diffusion implicit models,

    J. Song, C. Meng, and S. Ermon, “Denoising diffusion implicit models,” inProc. Int. Conf. Learning Representations (ICLR), Vienna, Austria (Virtual conference), May. 2021

    2021

  33. [33]

    BBDM: Image-to-image translation with Brownian bridge diffusion models,

    B. Li, K. Xue, B. Liu, and Y . -K. Lai, “BBDM: Image-to-image translation with Brownian bridge diffusion models,” inProc. IEEE Conf. Comput. Vis. Pattern Recog. (CVPR), Vancouver, BC, Canada, Jun. 2023, pp. 1952–1961

    2023

  34. [34]

    Diffusion bridge implicit models,

    K. Zheng, G. He, J. Chen, F. Bao, and J. Zhu, “Diffusion bridge implicit models,” inProc. Int. Conf. Learning Representations (ICLR), Singapore, Apr. 2025

    2025

  35. [35]

    Brownian bridge-based diffusion channel denoising for ISAC massive MIMO systems,

    S. Xu, J. Zhang, Y . Xu, C. Li, and L. Yang, “Brownian bridge-based diffusion channel denoising for ISAC massive MIMO systems,”IEEE Trans. Cogn. Commun. Netw., vol. 11, no. 2, pp. 764–776, Jan. 2025

    2025

  36. [36]

    Cell-free diffusion uplink with fronthaul noise adapting arbitrary fronthaul structure,

    Z. Honget al., “Cell-free diffusion uplink with fronthaul noise adapting arbitrary fronthaul structure,”IEEE Trans. Commun., vol. 74, pp. 3501– 3516, Jan. 2026

    2026

  37. [37]

    Physical channels and modulation,

    ETSI 3GPP, “Physical channels and modulation,” 3GPP TS 38.211, v.18.6.0, 2025–04 [Online]. Available: https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetail s.aspx?specificationId=3213

    2025

  38. [38]

    Wireless interference recognition with multimodal learning,

    P. Wang, K. Ma, Y . Bai, C. Sun, Z. Wang, and S. Chen, “Wireless interference recognition with multimodal learning,”IEEE Trans. Wireless Commun., vol. 23, no. 12, pp. 18576–18591, Dec. 2024

    2024

  39. [39]

    Complex ratio masking for monaural speech separation,

    D. S. Williamson, Y . Wang, and D. Wang, “Complex ratio masking for monaural speech separation,”IEEE Trans. Audio, Speech, Language Process., vol. 24, no. 3, pp. 483–492, Dec. 2016

    2016

  40. [40]

    U-Net: Convolutional net- works for biomedical image segmentation,

    O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional net- works for biomedical image segmentation,” inProc. Mid. Image Comput. Comput.-Assisted Intervention Conf. (MICCAI), Munich, Oct. 2015, pp. 234–241

    2015

  41. [41]

    Image super- resolution using very deep residual channel attention networks,

    Y . Zhang, K. Li, K. Li, L. Wang, B. Zhong, and Y . Fu, “Image super- resolution using very deep residual channel attention networks,” inProc. Eur . Conf. Comput. Vis. (ECCV), Munich, Germany, Sep. 2018, pp. 294– 310

    2018

  42. [42]

    Bilinear repre- sentation for language-based image editing using conditional generative adversarial networks,

    X. Mao, Y . Chen, Y . Li, T. Xiong, Y . He, and H. Xue, “Bilinear repre- sentation for language-based image editing using conditional generative adversarial networks,” inProc. IEEE Int. Conf. Acoust., Speech, Signal Process. (ICASSP), Brighton, UK, May. 2019, pp. 2047–2051

    2019

  43. [43]

    Score-based generative modeling through stochastic differential equations,

    Y . Song, J. Sohl-Dickstein, D. P. Kingma, A. Kumar, S. Ermon, and B. Poole, “Score-based generative modeling through stochastic differential equations,” inProc. Int. Conf. Learning Representations (ICLR), Vienna, Austria (Virtual conference), May. 2021

    2021

  44. [44]

    Attention is all you need,

    A. Vaswaniet al., “Attention is all you need,” inProc. Adv. Neural Inf. Process. Syst. (NeurIPS), Long Beach, CA, USA, Dec. 2017, pp. 34–45

    2017

  45. [45]

    DPM-solver: A fast ODE solver for diffusion probabilistic model sampling in around 10 steps,

    C. Lu, Y . Zhou, F. Bao, J. Chen, C. Li, and J. Zhu, “DPM-solver: A fast ODE solver for diffusion probabilistic model sampling in around 10 steps,” inProc. Adv. Neural Inf. Process. Syst. (NeurIPS), New Orleans, LA, USA, Dec. 2022, pp. 5775–5787

    2022

  46. [46]

    DemodNet: Learning soft demodulation from hard information using convolutional neural network,

    S. Zheng, X. Zhou, S. Chen, P. Qi, C. Lou, and X. Yang, “DemodNet: Learning soft demodulation from hard information using convolutional neural network,” inProc. IEEE Int. Conf. Commun. (ICC), Seoul, Korea, Republic of, May. 2022, pp. 1–6

    2022

  47. [47]

    Study on channel model for frequencies from 0.5 to 100 GHz,

    ETSI 3GPP, “Study on channel model for frequencies from 0.5 to 100 GHz,”3GPP TS 38.901, v.17.1.0, 2024–01 [Online]. Available: https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetail s.aspx?specificationId=3173

    2024

  48. [48]

    Conv-TasNet: Surpassing ideal time–frequency magnitude masking for speech separation,

    Y . Luo and N. Mesgarani, “Conv-TasNet: Surpassing ideal time–frequency magnitude masking for speech separation,”IEEE Trans. Audio, Speech, Language Process., vol. 27, no. 8, pp. 1256–1266, May. 2019

    2019