Pith. sign in

REVIEW 4 major objections 4 minor 53 references

Extract the Best, Discard the Rest: CSI Feedback with Offline Large AI Models

T0 review · 4 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Offline AI models sharpen MIMO feedback codebooks at zero latency.

desk verdict Plausible offline LVM codebook-refinement framework, but the vision-transfer claim is untested without a same-architecture random-init ablation. read the letter →

arxiv 2505.08566 v1 pith:7YYL3YY7 submitted 2025-05-13 eess.SP

Jialin Zhuang , Yafei Wang , Hongwei Hou , Yu Han , Wenjin Wang , Shi Jin , Jiangzhou Wang This is my paper · ORCID
classification eess.SP
keywords CSIfeedbackmassiveMIMOcodebookoptimizationlargevisionmodelofflineinferencefrequency-divisionduplexchannelstateinformationLVM4CF
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proposes a way to put large AI models into frequency-division duplex (FDD) massive MIMO channel feedback without paying their usual cost: instead of running a large model in real time on every channel vector, the model works offline to convert an ordinary RVQ codebook into an environment-tailored codebook. The refined codebook is then used at the base station, and optionally at the user, for the same low-bit feedback procedure, so the online path adds no inference latency or computational overhead. The paper claims that its model, LVM4CF, a vision-backbone network pre-trained on images and fine-tuned on CSI, attains higher cosine similarity between reconstructed and true channels and higher sum rate than conventional RVQ, FC, Transformer, and LLM baselines across site-specific and multi-scenario deployments. This matters because it offers a standard-compatible route to harvesting large-model representation power under tight latency constraints.

What carries the argument

The load-bearing mechanism is the offline codebook-refinement map $g_i = f_{\mathrm{LVM4CF}}(c_i;\Theta)$, which takes the $i$-th codeword of the conventional RVQ codebook and returns a unit-normalized refined codeword trained against the loss $\mathcal{L}_\rho = -\mathbb{E}_{\mathcal{H}_{\text{batch}}}\{|g^H h|/(\|g\|\|h\|)\}$. This map is built from a vision transformer pre-trained on image tokens and fine-tuned on CSI, exploiting the claimed structural analogy between spatial correlations in images and the dual-polarized antenna array correlations in CSI codewords. The same map powers both frameworks, and the two training algorithms differ only in whether the refined codebook is used at one side or shared and iteratively updated at both sides.

What would settle it

Train LVM4CF from scratch with the same architecture, data, and loss but with random initialization instead of image pre-training; if its cosine similarity and sum rate match the pre-trained version, then the image-transfer premise is not what carries the result.

Watch

Extended reading notes

Core claim

The central claim is that codebook-based CSI feedback can be treated as an offline codeword-refinement problem and that a large vision model is the right tool for it. LVM4CF takes each codeword from the conventional random-vector-quantization codebook, splits it into its two dual-polarization subvectors, embeds them, passes them through the pre-trained transformer layers, and outputs a refined codeword; the model is fine-tuned to maximize the expected cosine similarity between the refined codeword and the true channel. Two frameworks use this network: SSLCF fine-tunes it on site-specific CSI to produce one customized codebook, while MSLCF fine-tunes it on several environment types to produce a library of codebooks selected by an environment index. In dual-side mode the refined codebook is shared by base station and user and is iteratively updated during training; in single-side mode the user keeps the conventional codebook while the base station alone uses the refined one. The reported simulations show the refined codebooks improving both cosine similarity and downlink sum rate over the baselines, with the gain growing at high SNR.

Load-bearing premise

The load-bearing premise is that spatial correlations learned from natural images transfer to dual-polarized CSI codewords, so that image pre-training, rather than model size or training procedure, is what gives LVM4CF its edge.

Editorial extensions

If this is right

  • In single-side deployment, operators can improve CSI reconstruction while leaving user equipment on the existing RVQ codebook, preserving backward compatibility.
  • In dual-side deployment, the shared refined codebook yields the largest cosine similarity and sum-rate gains, at the cost of delivering the codebook once to users.
  • MSLCF reduces that delivery cost to a short environment index, making multi-scenario deployment feasible without retraining at each site.
  • Because all large-model computation happens offline, adding a large vision model does not increase online feedback running time, as the paper's timing comparison shows.
  • At high SNR, where interference dominates, the accuracy of the refined codebooks translates into widening sum-rate gains over conventional feedback.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not isolate pre-training from capacity: a randomly initialized LVM4CF trained on the same CSI data is not compared, so part of the reported gain could come from the architecture alone rather than from image transfer.
  • If the image-to-CSI transfer claim is right, the gain should scale with the diversity of the pre-training image data; a direct test would train variants on more or less visual data and track cosine similarity.
  • The dual-side iterative update can be viewed as a functional fixed-point search over codebooks; its convergence is empirically validated but not theoretically guaranteed, so a formal convergence analysis would be a natural next step.
  • The environment-index mechanism of MSLCF could be extended to online environment detection, letting the base station switch codebooks as the propagation environment changes.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper proposes two offline frameworks, SSLCF and MSLCF, for FDD massive MIMO CSI feedback. Instead of running a large model in real time, a large vision model called LVM4CF (a LLaMA-style backbone pre-trained on images and fine-tuned on CSI data) is used offline to refine each codeword of a conventional RVQ codebook into an environment-aware codebook. In the online stage the UE only feeds back the index of the best codeword and the BS looks it up, so no large-model inference is incurred. SSLCF produces site-specific codebooks, while MSLCF pre-generates a set of environment-specific codebooks and selects one with a low-overhead environment index. The paper details the LVM4CF architecture, two training and inference algorithms (single-side and dual-side), and simulations using QuaDRiGa channels showing cosine similarity and sum-rate gains over conventional codebook, FC, lightweight Transformer, and a GPT2-XL-based LLM baseline.

Significance. The offline-use idea is attractive: it reconciles the expressive power of large pretrained models with the latency constraints of wireless deployments, and the two deployment modes are clearly motivated. The paper is generally well structured, gives explicit algorithms, and evaluates on a standard simulator with open-source pretrained weights, which aids reproducibility. If the vision-transfer mechanism were substantiated, the result would be a useful design principle for codebook-based feedback. However, the current experiments do not isolate the effect of image pretraining from model capacity, and the dual-side training algorithm contains an ambiguity about parameter reinitialization; these issues need to be resolved before the central claim can be accepted.

major comments (4)
  1. [Section IV-B, Figs. 6-9] The central claim that image-domain pretraining transfers to CSI codeword refinement is not isolated by the experiments. LVM4CF is compared only against FC, a lightweight Transformer, and GPT2-XL, which differ in architecture, depth, parameter count, and pretraining data. Because no same-architecture randomly initialized baseline and no variant with only the CSI-specific embedding and output layers reinitialized are reported, the gains shown in Figs. 6, 7, 8(c), and 9 are also consistent with the simpler hypothesis that a larger model fine-tuned on CSI data yields better codebooks regardless of initialization. Since Section IV-B explicitly attributes the improvement to the structural analogy between images and CSI, an ablation separating pretrained initialization from model capacity is load-bearing and should be added.
  2. [Algorithm 2, line 8] The instruction 'Reinitialize the trainable parameters in LVM' is ambiguous and potentially undermines the transfer claim in dual-side deployment. If all trainable parameters are reset at each codebook update, the image-pretrained initialization is discarded whenever the codebook is refreshed, which contradicts the premise that pretrained knowledge is being exploited. If only the added heads are reset, the paper should say so explicitly. The surrounding text also says the update mechanism 'guarantees' consistent improvement, but the algorithm only accepts a codebook when the validation loss decreases; that is an empirical selection rule, not a guaranteed monotonic improvement.
  3. [Section V, baseline set] The comparison is restricted to codebook-refinement approaches; no end-to-end DL CSI feedback method such as CsiNet [15], CsiNet+ [19], or TransNet [22] is included. The abstract and introduction claim that the frameworks significantly outperform existing schemes, but the experiments only support superiority over the listed codebook-based baselines. Either add at least one representative end-to-end method under the same channel data and feedback-bit budget, or explicitly limit the claim to codebook-based feedback frameworks.
  4. [Section V, all figures] No error bars, multiple seeds, or confidence intervals are reported. The conclusion that LVM4CF consistently outperforms the LLM baseline, particularly the vision-versus-text comparison in Figs. 8(c) and 9, would be more convincing with at least three independent runs per configuration and a statement of variance.
minor comments (4)
  1. [References [4] and [33]] References [4] and [33] are assigned the same arXiv identifier (2406.09022), which appears to be an error; the LLM-CSI-feedback reference should be corrected.
  2. [Section V, LLM baseline] The LLM baseline is described as a '12-layer GPT2-XL,' but GPT-2 XL has 48 layers; if a 12-layer GPT-2 variant was used, it should be named and configured accordingly.
  3. [Title, Section I, Section V, Notation] Typos such as 'multiple-intput' in the title and abstract, 'practicale propagation environments' in Section I, 'choose' for 'chosen' in Section V, and 'thei-th toj-th' in the Notation paragraph should be corrected.
  4. [Fig. 8(b)] Please clarify whether the reported running time includes only the online codebook lookup and index feedback or also the offline codebook-generation time; the 'no added latency' claim depends on this distinction.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the reported codebook gains are evaluated on held-out CSI data and do not reduce to the training objective by construction.

full rationale

The paper's claimed derivation is an empirical training-and-evaluation pipeline rather than a closed-form argument that equates outputs with inputs. LVM4CF is fine-tuned by minimizing the cosine-similarity loss in Eq. (25) on training pairs, while the reported similarities in Figs. 5-9 are computed on separate test samples (e.g., 70k/20k/10k splits for SSLCF and 140k/40k/20k for MSLCF in Section V). Because the test channels are not used to fit the network or select the final codebook, the claim that refined codewords improve cosine similarity is not forced by construction. The DS training loop in Algorithm 2 updates the codebook only when validation loss improves, but the final comparison is against held-out test channels, which is standard model selection rather than circular prediction. The vision-to-CSI transfer argument in Section IV-B is an empirical assumption about structural analogy, and the paper does not include a same-architecture randomly initialized ablation; that is a legitimate experimental gap relevant to attribution and correctness, but it is not circularity, because the paper does not define 'transfer' in terms of the results it later reports. Self-citations such as [25], [33], and [34] appear as related work and baselines only, and none is invoked as a theorem or uniqueness result that forces the architecture or the numerical outcome. The simulations are anchored to the externally validated QuaDRiGa channel simulator, and the central comparison is against standard RVQ, FC, Transformer, and LLM baselines on held-out data. Overall, no load-bearing step reduces to its own inputs.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim rests on two domain assumptions: channel statistics are stable per site, and image pre-training transfers to CSI codeword refinement. No ad hoc free parameters or invented physical entities are introduced; hyperparameters are standard training choices.

assumptions (4)
  • domain assumption Wireless propagation environment within a BS coverage area remains statistically stable over long timescales.
    Stated in Section III-A, this justifies offline site-specific codebook generation.
  • domain assumption Spatial correlations in images are structurally analogous to those in dual-polarized CSI codewords, enabling transfer of pre-trained vision representations.
    Stated in Section IV-B; asserted without empirical isolation from model capacity.
  • domain assumption QuaDRiGa is a reliable channel model for the studied UMa and RMa scenarios.
    Used as ground truth for all simulations and stated in Section V as validated by 3GPP.
  • standard math ZF precoding and the sum-rate formula correctly represent system performance.
    Standard formulations used in Section II, not questioned by the authors.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Extract the Best, Discard the Rest: CSI Feedback with Offline Large AI Models." pith.science (2026). https://pith.science/paper/7YYL3YY7

@misc{pith2026250508566,
  author       = {Pith},
  title        = {Pith review of: Extract the Best, Discard the Rest: CSI Feedback with Offline Large AI Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7YYL3YY7}},
  note         = {Machine review of arXiv:2505.08566}
}
read the original abstract

Large AI models (LAMs) have shown strong potential in wireless communication tasks, but their practical deployment remains hindered by latency and computational constraints. In this work, we focus on the challenge of integrating LAMs into channel state information (CSI) feedback for frequency-division duplex (FDD) massive multiple-intput multiple-output (MIMO) systems. To this end, we propose two offline frameworks, namely site-specific LAM-enhanced CSI feedback (SSLCF) and multi-scenario LAM-enhanced CSI feedback (MSLCF), that incorporate LAMs into the codebook-based CSI feedback paradigm without requiring real-time inference. Specifically, SSLCF generates a site-specific enhanced codebook through fine-tuning on locally collected CSI data, while MSLCF improves generalization by pre-generating a set of environment-aware codebooks. Both of these frameworks build upon the LAM with vision-based backbone, which is pre-trained on large-scale image datasets and fine-tuned with CSI data to generate customized codebooks. This resulting network named LVM4CF captures the structural similarity between CSI and image, allowing the LAM to refine codewords tailored to the specific environments. To optimize the codebook refinement capability of LVM4CF under both single- and dual-side deployment modes, we further propose corresponding training and inference algorithms. Simulation results show that our frameworks significantly outperform existing schemes in both reconstruction accuracy and system throughput, without introducing additional inference latency or computational overhead. These results also support the core design methodology of our proposed frameworks, extracting the best and discarding the rest, as a promising pathway for integrating LAMs into future wireless systems.

Figures

Figures reproduced from arXiv: 2505.08566 by the authors.

Figure 1
Figure 1. Example of deployment scenario with conventional and environment [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. The two proposed codebook-based CSI feedback frameworks. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Network architectures of LVM4CF: Vision task vs. CSI codebook enhancement. [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Training strategy for generating enhanced codebook. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 6
Figure 6. Figure 6: Cosine similarity vs. the number of feedback bits [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 5
Figure 5. Figure 5: Cosine similarity vs. the number of feedback bits [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 8
Figure 8. Figure 8: Comparison of model complexity, running time, and communication [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]
Figure 9
Figure 9. Figure 9: Sum rate vs. SNR (in UMa-mmWave scenario). [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

53 extracted references · 40 canonical work pages

  1. [34]

    AI-enhanced codebook-based CSI feedback in FDD massive MIMO,

    J. Guo, C.-K. Wen, M. Chen, and S. Jin, “AI-enhanced codebook-based CSI feedback in FDD massive MIMO,” in 2021 IEEE 94th Vehicular Technology Conference (VTC2021-Fall), 2021, pp. 1–5

    work page 2021

  2. [33]

    Exploring the potential of large language models for massive MIMO CSI feedback,

    Y . Cui, J. Guo, C.-K. Wen, S. Jin, and E. Tong, “Exploring the potential of large language models for massive MIMO CSI feedback,” arXiv:2406.09022

    arXiv

  3. [15]

    Deep learning for massive MIMO CSI feedback,

    C.-K. Wen, W.-T. Shih, and S. Jin, “Deep learning for massive MIMO CSI feedback,” IEEE Wirel. Commun. Lett. , vol. 7, no. 5, pp. 748–751, Oct. 2018

    work page 2018

  4. [19]

    Convolutional neural network- based multiple-rate compressive sensing for massive MIMO CSI feed- back: Design, simulation, and analysis,

    J. Guo, C.-K. Wen, S. Jin, and G. Y . Li, “Convolutional neural network- based multiple-rate compressive sensing for massive MIMO CSI feed- back: Design, simulation, and analysis,” IEEE Trans. Wireless Commun., vol. 19, no. 4, pp. 2827–2840, Jan. 2020

    work page 2020

  5. [22]

    Transnet: Full attention network for CSI feedback in FDD massive MIMO system,

    Y . Cui, A. Guo, and C. Song, “Transnet: Full attention network for CSI feedback in FDD massive MIMO system,” IEEE Wireless Commun. Lett., vol. 11, no. 5, pp. 903–907, May 2022

    work page 2022

  6. [1]

    A tutorial on extremely large-scale MIMO for 6G: Fundamentals, signal processing, and applications,

    Z. Wang, J. Zhang, H. Du, D. Niyato, S. Cui, B. Ai, M. Debbah, K. B. Letaief, and H. V . Poor, “A tutorial on extremely large-scale MIMO for 6G: Fundamentals, signal processing, and applications,” IEEE Commun. Surv. Tutorials, vol. 26, no. 3, pp. 1560–1605, Jan. 2024

    work page 2024

  7. [2]

    Spatial- and frequency- wideband effects in millimeter-wave massive MIMO systems,

    B. Wang, F. Gao, S. Jin, H. Lin, and G. Y . Li, “Spatial- and frequency- wideband effects in millimeter-wave massive MIMO systems,” IEEE Trans. Signal Process., vol. 66, no. 13, pp. 3393–3406, Jul. 2018

    work page 2018

  8. [3]

    An overview of massive MIMO: Benefits and challenges,

    L. Lu, G. Y . Li, A. L. Swindlehurst, A. Ashikhmin, and R. Zhang, “An overview of massive MIMO: Benefits and challenges,” IEEE J. Sel. Topics Signal Process., vol. 8, no. 5, pp. 742–758, Oct. 2014

    2014

Show all 53 references
  1. [5]

    A tensor-structured approach to dynamic channel prediction for massive MIMO systems with temporal non-stationarity,

    H. Hou, Y . Wang, Y . Zhu, X. Yi, W. Wang, D. T. M. Slock, and S. Jin, “A tensor-structured approach to dynamic channel prediction for massive MIMO systems with temporal non-stationarity,” arXiv:2412.06713

    arXiv

  2. [6]

    On the performance of channel-statistics-based codebook for massive MIMO channel feed- back,

    W. Shen, L. Dai, Y . Zhang, J. Li, and Z. Wang, “On the performance of channel-statistics-based codebook for massive MIMO channel feed- back,” IEEE Trans. Veh. Technol., vol. 8, no. 66, pp. 7553–7557, Aug. 2017

    2017

  3. [7]

    5G new radio (NR): Physical layer overview and perfor- mance,

    A. Ghosh, “5G new radio (NR): Physical layer overview and perfor- mance,” in Proc. IEEE Commun. Theory Workshop , 2018, pp. 1–38

    2018

  4. [8]

    An overview of limited feedback in wireless commu- nication systems,

    D. J. Love, R. W. Heath, V . K. N. Lau, D. Gesbert, B. D. Rao, and M. Andrews, “An overview of limited feedback in wireless commu- nication systems,” IEEE J. Sel. Areas Commun. , vol. 26, no. 8, pp. 1341–1365, Oct. 2008

    2008

  5. [9]

    Spatially common sparsity based adaptive channel estimation and feedback for FDD massive MIMO,

    Z. Gao, L. Dai, Z. Wang, and S. Chen, “Spatially common sparsity based adaptive channel estimation and feedback for FDD massive MIMO,” IEEE Trans. Signal Process., vol. 63, no. 23, pp. 6169–6183, Dec. 2012

    2012

  6. [10]

    An iterative thresholding al- gorithm for linear inverse problems with a sparsity constraint,

    I. Daubechies, M. Defrise, and C. De Mol, “An iterative thresholding al- gorithm for linear inverse problems with a sparsity constraint,” Commun. Pure Appl. Math. , vol. 57, no. 11, pp. 1413–1457, 2004

    2004

  7. [11]

    From denoising to compressed sensing,

    C. A. Metzler, A. Maleki, and R. G. Baraniuk, “From denoising to compressed sensing,” IEEE Trans. Inf. Theory., vol. 62, no. 9, pp. 5117– 5144, Sep. 2016

    2016

  8. [12]

    Distributed compressive CSIT estimation and feedback for FDD multi-user massive MIMO systems,

    X. Rao and V . K. N. Lau, “Distributed compressive CSIT estimation and feedback for FDD multi-user massive MIMO systems,” IEEE Trans. Signal Process., vol. 62, no. 12, pp. 3261–3271, Jun. 2014

    2014

  9. [13]

    Compressive sensing for feedback reduction in MIMO broadcast channels,

    M. E. Eltayeb, T. Y . Al-Naffouri, and H. R. Bahrami, “Compressive sensing for feedback reduction in MIMO broadcast channels,” IEEE Trans. Commun., vol. 62, no. 9, pp. 3209–3222, Sep. 2014

    2014

  10. [14]

    Sparse representation for wireless communications: A compressive sensing approach,

    Z. Qin, J. Fan, Y . Liu, Y . Gao, and G. Y . Li, “Sparse representation for wireless communications: A compressive sensing approach,” IEEE Trans. Signal Process., vol. 35, no. 3, pp. 40–58, May 2018

    2018

  11. [16]

    Deep learning-based CSI feedback approach for time-varying massive MIMO channels,

    T. Wang, C.-K. Wen, S. Jin, and G. Y . Li, “Deep learning-based CSI feedback approach for time-varying massive MIMO channels,” IEEE Wirel. Commun. Lett., vol. 8, no. 2, pp. 416–419, Aug. 2019

    2019

  12. [17]

    CsiNet-former network for bilateral user 3D localization in STAR-RIS-assisted MISO systems,

    J. Li, W. Wang, R. Jiang, and S. Huang, “CsiNet-former network for bilateral user 3D localization in STAR-RIS-assisted MISO systems,” in 2023 IEEE Globecom Workshops (GC Wkshps) , 2023, pp. 1886–1891

    2023

  13. [18]

    Auto-CsiNet: Scenario-customized automatic neural network architecture generation for massive MIMO CSI feedback,

    X. Li, J. Guo, C.-K. Wen, and S. Jin, “Auto-CsiNet: Scenario-customized automatic neural network architecture generation for massive MIMO CSI feedback,” IEEE Trans. Wireless Commun. , vol. 23, no. 10, pp. 14 759–14 775, Oct. 2024

    2024

  14. [20]

    Attention model for massive MIMO CSI compression feedback and recovery,

    Q. Cai, C. Dong, and K. Niu, “Attention model for massive MIMO CSI compression feedback and recovery,” in 2019 IEEE Wireless Communi- cations and Networking Conference (WCNC) , 2019, pp. 1–5

    2019

  15. [21]

    Transformer empowered CSI feed- back for massive MIMO systems,

    Y . Xu, M. Yuan, and M.-O. Pun, “Transformer empowered CSI feed- back for massive MIMO systems,” in 2021 30th Wireless and Optical Communications Conference (WOCC) , 2021, pp. 157–161

    2021

  16. [23]

    CAnet: Uplink-aided downlink channel acquisition in FDD massive MIMO using deep learning,

    J. Guo, C.-K. Wen, and S. Jin, “CAnet: Uplink-aided downlink channel acquisition in FDD massive MIMO using deep learning,” IEEE Trans. Signal Process., vol. 70, no. 1, pp. 199–214, Jan. 2022

    2022

  17. [24]

    Exploiting bi-directional channel reciprocity in deep learning for low rate massive MIMO CSI feedback,

    Z. Liu, L. Zhang, and Z. Ding, “Exploiting bi-directional channel reciprocity in deep learning for low rate massive MIMO CSI feedback,” IEEE Wireless Commun. Lett. , vol. 8, no. 3, pp. 889–892, Jun. 2019

    2019

  18. [25]

    CovNet: Covariance information-assisted CSI feedback for FDD massive MIMO systems,

    J. Zhuang, X. He, Y . Wang, J. Liu, and W. Wang, “CovNet: Covariance information-assisted CSI feedback for FDD massive MIMO systems,” IEEE Wireless Commun. Lett. , vol. 14, no. 3, pp. 641–645, Mar. 2025

    2025

  19. [26]

    Bit-level optimized neural network for multi-antenna channel quantization,

    C. Lu, W. Xu, S. Jin, and K. Wang, “Bit-level optimized neural network for multi-antenna channel quantization,” IEEE Wireless Commun. Lett. , vol. 9, no. 1, pp. 87–90, Jan. 2020

    2020

  20. [27]

    Binary neural network aided CSI feedback in massive MIMO system,

    Z. Lu, J. Wang, and J. Song, “Binary neural network aided CSI feedback in massive MIMO system,” IEEE Wireless Commun. Lett., vol. 10, no. 6, pp. 1305–1308, Jun. 2021

    2021

  21. [28]

    Multi-resolution CSI feedback with deep learning in massive MIMO system,

    ——, “Multi-resolution CSI feedback with deep learning in massive MIMO system,” in ICC 2020 - 2020 IEEE International Conference on Communications (ICC), 2020, pp. 1–6

    2020

  22. [29]

    Attention is all you need,

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, and I. Polosukhin, “Attention is all you need,” in Proc. Adv. Neural Inf. Process. Syst. (NIPS) , Dec. 2017, pp. 5998–6008

    2017

  23. [30]

    Language models are unsupervised multitask learners,

    A. Radford, J. Wu, R. Child, D. Luan, D. Amodei, and I. Sutskever, “Language models are unsupervised multitask learners,” OpenAI blog , vol. 1, no. 8, p. 9, 2019

    2019

  24. [31]

    Large language model enhanced multi-agent systems for 6G communications,

    F. Jiang, Y . Peng, L. Dong, K. Wang, K. Yang, C. Pan, D. Niyato, and O. A. Dobre, “Large language model enhanced multi-agent systems for 6G communications,” IEEE Wireless Commun. , vol. 31, no. 6, pp. 48–55, Dec. 2024

    2024

  25. [32]

    Large AI model-based semantic communications,

    F. Jiang, Y . Peng, L. Dong, K. Wang, K. Yang, C. Pan, and X. You, “Large AI model-based semantic communications,” IEEE Wireless Com- mun., vol. 31, no. 3, pp. 68–75, Jun. 2024

    2024

  26. [35]

    Linear transmit processing in MIMO communications systems,

    M. Joham, W. Utschick, and J. Nossek, “Linear transmit processing in MIMO communications systems,” IEEE Trans. Signal Process., vol. 53, no. 8, pp. 2700–2712, Jul. 2005

    2005

  27. [36]

    A vector-perturbation technique for near-capacity multiantenna multiuser communication-part i: channel inversion and regularization,

    C. Peel, B. Hochwald, and A. Swindlehurst, “A vector-perturbation technique for near-capacity multiantenna multiuser communication-part i: channel inversion and regularization,” IEEE Trans. Commun., vol. 53, no. 1, pp. 195–202, Feb. 2005

    2005

  28. [37]

    Weighted sum-rate maximization using weighted MMSE for MIMO- BC beamforming design,

    S. S. Christensen, R. Agarwal, E. De Carvalho, and J. M. Cioffi, “Weighted sum-rate maximization using weighted MMSE for MIMO- BC beamforming design,” IEEE Trans. Wireless Commun., vol. 7, no. 12, pp. 4792–4799, Dec. 2008

    2008

  29. [38]

    Overview of deep learning- based CSI feedback in massive MIMO systems,

    J. Guo, C.-K. Wen, S. Jin, and G. Y . Li, “Overview of deep learning- based CSI feedback in massive MIMO systems,” IEEE Trans. Commun., vol. 70, no. 12, pp. 8017–8045, Dec. 2022

    2022

  30. [39]

    A review of codebooks for CSI feedback in 5G new radio and beyond,

    Z. Qin and H. Yin, “A review of codebooks for CSI feedback in 5G new radio and beyond,” arXiv:2302.09222

    arXiv

  31. [40]

    NR; physical layer proce- dures for data,

    3rd Generation Partnership Project (3GPP), “NR; physical layer proce- dures for data,” 3GPP, TS 38.214, V .18.6.0, 2025

    2025

  32. [41]

    Beam-delay domain channel estimation for mmWave XL-MIMO systems,

    H. Hou, X. He, T. Fang, X. Yi, W. Wang, and S. Jin, “Beam-delay domain channel estimation for mmWave XL-MIMO systems,” IEEE J. Sel. Topics Signal Process. , vol. 18, no. 4, pp. 646–661, May. 2024. 13

    2024

  33. [42]

    Site- specific online compressive beam codebook learning in mmwave vehic- ular communication,

    Y . Wang, N. J. Myers, N. Gonz ´alez-Prelcic, and R. W. Heath, “Site- specific online compressive beam codebook learning in mmwave vehic- ular communication,” IEEE Trans. Wireless Commun. , vol. 20, no. 5, pp. 3122–3136, May. 2021

    2021

  34. [43]

    LWM: A pre-trained wireless foundation model for universal feature extraction,

    S. Alikhani, G. Charan, and A. Alkhateeb, “LWM: A pre-trained wireless foundation model for universal feature extraction,” in Proc. IEEE Int. Conf. Mach. Learn. Commun. Netw. (ICMLCN). IEEE, 2025

    2025

  35. [44]

    NR; radio resource control (RRC); protocol specification,

    3rd Generation Partnership Project (3GPP), “NR; radio resource control (RRC); protocol specification,” 3GPP, TS 38.331, V .18.5.1, 2025

    2025

  36. [45]

    Gaussian error linear units (GELUs),

    D. Hendrycks and K. Gimpel, “Gaussian error linear units (GELUs),” 2023, arXiv:1606.08415

    2023 arXiv

  37. [46]

    Layer normalization,

    J. L. Ba, J. R. Kiros, and G. E. Hinton, “Layer normalization,” 2016, arXiv:1607.06450

    2016 arXiv

  38. [47]

    Roformer: En- hanced transformer with rotary position embedding,

    J. Su, Y . Lu, S. Pan, A. Murtadha, B. Wen, and Y . Liu, “Roformer: En- hanced transformer with rotary position embedding,” arXiv:2104.09864

    arXiv

  39. [48]

    Root mean square layer normalization,

    B. Zhang and R. Sennrich, “Root mean square layer normalization,” 2019, arXiv:1910.07467

    2019 arXiv

  40. [49]

    Activation functions in deep learning: A comprehensive survey and benchmark,

    S. R. Dubey, S. K. Singh, and B. B. Chaudhuri, “Activation functions in deep learning: A comprehensive survey and benchmark,” 2022, arXiv:2109.14545

    2022 arXiv

  41. [50]

    Sequential modeling enables scalable learning for large vision models,

    Y . Bai, X. Geng, K. Mangalam, A. Bar, A. L. Yuille, T. Darrell, J. Malik, and A. A. Efros, “Sequential modeling enables scalable learning for large vision models,” in 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 22 861–22 872

    2024

  42. [51]

    Taming transformers for high- resolution image synthesis,

    P. Esser, R. Rombach, and B. Ommer, “Taming transformers for high- resolution image synthesis,” in 2021 IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR) , Jun. 2021, pp. 12 868– 12 878

    2021

  43. [52]

    QuaDRiGa: A 3- d multi-cell channel model with time evolution for enabling virtual field trials,

    S. Jaeckel, L. Raschkowski, K. B ¨orner, and L. Thiele, “QuaDRiGa: A 3- d multi-cell channel model with time evolution for enabling virtual field trials,” IEEE Trans. Antennas Propag. , vol. 62, no. 6, pp. 3242–3256, Jun. 2014

    2014

  44. [53]

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

    3rd Generation Partnership Project (3GPP), “Study on channel model for frequencies from 0.5 to 100 GHz,” 3GPP, TR 38.901, V .18.0.0, 2024

    2024

  45. [54]

    MIMO broadcast channels with finite-rate feedback,

    N. Jindal, “MIMO broadcast channels with finite-rate feedback,” IEEE Trans. Inf. Theory., vol. 52, no. 11, pp. 5045–5060, Nov. 2006

    2006

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.