Pith. sign in

REVIEW 4 major objections 4 minor 5 cited by

Temporal Spectrum Cartography in Low-Altitude Economy Networks: A Generative AI Framework with Multi-Agent Learning

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A two-stage generative framework—a masked autoencoder plus a multi-agent diffusion policy—reconstructs time-varying radio spectrum maps from sparse UAV and static sensor data, cutting reconstruction error by 57.35% versus Kriging and…

desk verdict A well-written integration of MAE reconstruction and diffusion-based multi-agent planning, undermined by a central objective that, as written, punishes the planner for predicting signal in unsensed areas. read the letter →

arxiv 2505.15571 v1 pith:2GNJ7ITK submitted 2025-05-21 eess.SP

Changyuan Zhao , Ruichen Zhang , Jiacheng Wang , Dusit Niyato , Geng Sun , Hongyang Du , Zan Li , Abbas Jamalipour
show 1 more author
Dong In Kim
This is my paper · ORCID
classification eess.SP
keywords temporalspectrumcartographylow-altitudeeconomynetworksmaskedautoencodermulti-agentdiffusionpolicysparseradiomapreconstructionUAVtrajectoryoptimizationgenerativeAIforwirelessreinforcementlearning
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 tries to establish that temporal spectrum cartography in low-altitude economy networks can be treated as a two-stage generative problem rather than a static interpolation problem. In the first stage, a reconstructive masked autoencoder uses a dual-mask mechanism to reconstruct complete time-varying radio power maps from sparse static and UAV sensor measurements. In the second stage, a multi-agent diffusion policy plans UAV trajectories to minimize cumulative reconstruction error over time. If the claims hold, operators could map dynamic radio environments from a handful of aerial sensors and move sensing assets where they reduce map error most.

What carries the argument

The load-bearing objects are the two generative modules. RecMAE is a masked autoencoder that masks sensor data twice: a pixel-level mask drops individual measurements to mimic sparse sensing, and a patch-level mask hides whole spatio-temporal tubelets before the encoder, forcing the decoder to recover fine local detail and global context together. MADP is a multi-agent diffusion policy in which each UAV's actor is a conditional denoising diffusion model that generates actions by refining noise under a temporal-attention state encoder, trained with centralized critics and decentralized execution to encourage cooperation. Together they close the loop: the reconstructor turns sparse readings into maps, and the planner uses reconstruction quality to decide where the UAVs should sense next.

What would settle it

Run the same two-stage system in a field test with real UAV-collected RSSI over a known urban area and compare per-time-slot reconstructed maps against Kriging and autoencoder baselines; if RecMAE's error margin shrinks or reverses, the simulator-to-reality gap is the cause. Separately, retrain RecMAE with contiguous circular sensing masks instead of random pixel dropout and observe whether the reported advantage over baselines persists.

Watch

Extended reading notes

Core claim

The paper's central claim is that combining a generative reconstructor with a generative trajectory planner outperforms both classical interpolation and deep-learning baselines in dynamic low-altitude scenarios. Specifically, RecMAE is reported to cut reconstruction error by 57.35% against Kriging and 88.68% against an autoencoder across sensing ratios from 10% down to 3%, with the lowest standard deviation and no retuning between coverage levels. MADP is reported to reduce cumulative reconstruction MSE to 50.00 from 153.91 for a CNN-based multi-agent planner and from 95.04 for the attention-augmented variant, and its learned trajectories disperse UAVs across sub-regions to avoid redundant coverage. The paper also claims that the framework degrades gracefully when static sensors are removed or UAV team size shrinks, but performs best with both asset types present.

Load-bearing premise

The framework is trained and tested only on radio maps simulated from a standardized urban propagation model, with random pixel dropout standing in for real sensor coverage; if either the simulator or the mask model diverges from real low-altitude environments, the reported error reductions may not transfer.

Editorial extensions

If this is right

  • At 10% sensing coverage RecMAE reports MSE 0.39 versus 0.54 for Kriging and 0.56 for the autoencoder; at 3% it reports 0.90 versus 2.11 and 7.95, with a lower standard deviation than either baseline.
  • A model trained at 10% coverage is evaluated at 5% and 3% without modification, and the accuracy advantage persists, indicating robustness to sparser sensor deployments.
  • MADP's learned trajectories lower cumulative reconstruction MSE from 153.91 to 50.00 against the CNN planner and from 95.04 to 50.00 against the attention-augmented planner, with more stable training rewards.
  • The planner remains functional with no static sensors (cumulative MSE 225.71), and increasing UAV count from one to four reduces cumulative MSE from 495.60 to 50.00, so cooperation among mobile sensors is a direct source of accuracy.
  • RecMAE inference takes about 25 seconds for the full test set, roughly four times the autoencoder, but Kriging takes about 10^4 seconds, so the accuracy gain avoids Kriging's prohibitive runtime.

Reading between the lines

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

  • A testable extension is replacing random pixel dropout with contiguous sensing-footprint masks (real UAV and static sensors observe disk-shaped regions), which would show whether RecMAE's advantage survives realistic observation patterns.
  • The dual-mask design is generic enough to transfer to other sparse-sensing reconstruction tasks, such as traffic or environmental monitoring, where missing data is scattered at both fine and coarse scales.
  • The paper leaves open whether the diffusion planner's stochastic action generation improves performance beyond the attention mechanism itself; ablating the diffusion actor against a deterministic actor with the same encoder would isolate that contribution.
  • Field validation is the natural next step: the synthetic channel model's realism is the main uncertainty, and real RSSI experiments would determine whether the reported margins persist.
Share X Bluesky LinkedIn Reddit HN

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 a two-stage generative-AI framework for temporal spectrum cartography in low-altitude economy networks. Stage one, RecMAE, is a masked-autoencoder-style reconstructor with a dual-mask mechanism (pixel-level and patch-level) that is trained on sparse, temporally stacked radio maps. Stage two, MADP, is a multi-agent diffusion-policy planner, trained with centralized critics and decentralized execution, that chooses the next positions of dynamic UAV sensors so as to minimize cumulative reconstruction error. Experiments are run on a synthetic 3GPP TR 38.901 urban scenario. The authors report that RecMAE reduces reconstruction MSE by 57.35% relative to Kriging and 88.68% relative to an autoencoder baseline at 3% sensing ratio, and that MADP lowers cumulative MSE by 67.51% relative to a CNN-based MADDPG baseline. The central claim is that the integrated framework improves both reconstruction accuracy and sensor-trajectory efficiency for temporal spectrum mapping.

Significance. If the stated results hold, the paper would make a useful advance by coupling a generative reconstructor with a learned multi-agent planner for time-varying radio maps. The system model is clearly described, the channel-level simulation is standard, and the comparison to AE, Kriging, and CNN/CNN-Attention baselines is a reasonable first evaluation. The paper also reports runtime costs, which is helpful. However, the significance is currently limited by three factors: the formal planning objective in Eqs. (11)-(13) is misspecified as written and appears to penalize nonzero reconstructions in unsensed regions; Algorithm 1's patch-masking procedure is internally inconsistent and applies random masking at inference; and all conclusions rest on one synthetic simulator with no code or data release and no ablation of the claimed dual-mask contribution. These issues do not necessarily invalidate the reconstructor results, but they do mean that the planner results, and the attribution of gains to the proposed components, are not yet established.

major comments (4)
  1. [Section 3.3, Eqs. (11)-(13)] The reconstruction error is defined as ||tilde P_t - hat P_t||_2 where tilde P_t = W_t o P_t is zero outside the sensing footprint. Consequently, the objective in Eq. (13) contains a term ||(1-W_t) o hat P_t||^2 that penalizes any nonzero predicted power in unsensed cells. This is not the stated goal of recovering the complete radio map; it actively rewards predicting zero power wherever no sensor is located. If this quantity is used literally as the MADP reward in Eq. (34), the cumulative-MSE reductions in Figs. 6-8 may reflect optimization of this misspecified objective rather than improved true-map accuracy. Please replace tilde P_t with the true map P_t in Eqs. (11) and (13), or restrict the error to observed entries, e.g., ||W_t o (P_t - hat P_t)||_F^2, and rerun or clarify which objective the experiments actually used.
  2. [Section 4.3.3, Eq. (34), and Section 5.2] The MARL objective in Eq. (27) maximizes expected cumulative reward, but Eq. (34) defines R as the reconstruction error, and Section 5.2 states that the reward is '30 minus the error.' As printed, maximizing the reward in Eq. (34) would maximize reconstruction error. This is a formal contradiction with the stated goal of minimizing cumulative MSE. Please restate the reward as a decreasing function of reconstruction error, e.g., r = C - E_t, and make Eq. (34) consistent with the experiments.
  3. [Section 4.1.6 and Algorithm 1] The dual-mask procedure is not implemented consistently in the pseudocode. In Procedure 1, the encoder is applied to X_pixel before the patch-level mask indices V are generated, so the patch mask cannot affect the encoded representation. In Procedure 2, a random patch-level mask is generated at inference, which randomly discards a subset of the already-sparse sensor observations and makes the reconstruction stochastic; this also conflicts with the description in Section 4.1.8, where the input is only the sensor-induced mask W_i. Please correct the order of operations in training, remove or justify the random masking at inference, and report how the randomness is controlled in Tables 2 and Figs. 5-9.
  4. [Section 5, experimental support] All results are obtained on a single synthetic 3GPP TR 38.901 scenario with random pixel dropout as the proxy for sensor coverage. Because the planner reward is the reconstruction error of the authors' own RecMAE, the reported planner gains could partly reflect optimization against RecMAE-specific biases rather than true-map accuracy. In addition, there is no ablation isolating the pixel-level mask and patch-level mask, so the improvement over AE and Kriging cannot be attributed to the dual-mask design. Please add an ablation (patch-only, pixel-only, dual), evaluate the planner with at least one independent reconstruction surrogate, and, if possible, validate on a second simulator or real-measurement data. Releasing code and data, or at least fixed seeds, would materially help reproducibility.
minor comments (4)
  1. [Section 4.1.5] The sentence 'Note that tokens with indices in M are invisible tokens and are the input to the encoder' contradicts the previous sentence, which correctly states that only visible tokens V are fed to the encoder; 'M' should read 'V'.
  2. [Algorithm 2 and Eq. (37)] The actor notation is inconsistent: the input lists actor networks {mu_theta_i}, but the loop and target-action lines use pi'_theta and pi_theta, and the actor loss in Eq. (37) has mismatched parentheses. Please unify the notation and correct the loss expression.
  3. [Section 4.2.1 and Eq. (34)] The POMDP description indexes time slots as T1,...,T_nt, while Eq. (34) sums from t=0 to nT-1; please align the time indexing throughout.
  4. [Fig. 6(b) and Section 5.2] The text reports a Random-policy cumulative MSE of 361.55, while Fig. 6(b) shows 361.77; please reconcile the numbers.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: RecMAE is evaluated against ground-truth maps and external baselines, and the MADP-vs-baseline comparison uses a shared reward; the main caveat is a reward-proxy mismatch.

full rationale

The reconstruction-stage claim is self-contained: RecMAE's training loss (Eq. 26) compares reconstructed patches against ground truth, and Table 2 and Fig. 5 report MSE against the true radio power maps, with AE and Kriging as external baselines, so the 88.68% and 57.35% reductions are not defined into existence. The planning stage defines the shared reward as 30 minus the reconstruction error (Section 5.2), and Eq. (34) identifies that error with the GenAI reconstructor's output; because CNN, CNN-Attention, and Random baselines are optimized and evaluated under the same reward, MADP's higher reward and lower cumulative MSE in Figs. 6-8 are an empirical algorithm comparison, not a tautology. The relevant non-circular concern is that Eqs. (10)-(11) and (13) define reconstruction error against the sparse sensed map \tilde P_t = W_t \circ P_t rather than the true map P_t, so the planner reward may reward zero predictions in unsensed regions and may not track true-map accuracy; this is a validity and objective-mismatch risk, not a circular reduction. Self-citations (e.g., [10], [13], [22], [30], [31]) are contextual background and are not load-bearing for the core derivation.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The framework adds no new physical entities. The performance claims rest on a synthetic dataset generated from a standard 3GPP channel model, hand-selected hyperparameters, and the untested assumption that random pixel masks represent real sensor footprints. These choices are reasonable simulation assumptions but they are not externally validated.

free parameters (6)
  • patch masking ratio rpatch = 0.75
    Chosen by hand in Table 1; controls the fraction of patches hidden from the encoder. No sensitivity analysis is reported.
  • pixel masking ratio rpixel = 0.90
    Set to 0.90 in Table 1 to mimic 10% sensing coverage during training; the model is then tested at 5% and 3% sensing ratios.
  • reward offset = 30
    Reward is defined as 30 minus reconstruction error in Section 5.2; the constant is arbitrary and not tuned or justified.
  • denoising steps T = 6
    Number of reverse diffusion steps in MADP; chosen in Section 5.2 without an ablation.
  • sensing footprints Rd and Rs = 3x3 and 1x1 grid cells
    Dynamic UAVs sense a 3x3 neighborhood, static sensors a 1x1 point; these assumptions determine the pixel masks at deployment.
  • movement limit dm = 2 grid cells
    UAVs can move up to two cells per time slot; this restricts the action space and is not varied or justified.
assumptions (6)
  • domain assumption 3GPP TR 38.901 urban propagation model
    Power maps are generated from LOS/NLOS path loss in Eqs. (2)-(7); all conclusions inherit the simulator's assumptions.
  • domain assumption Exponentially decaying spatial shadowing correlation
    Eq. (6) assumes Cov = sigma^2 exp(-d/d_corr), a standard but specific model.
  • domain assumption Sensor mask constancy within each time slot
    Eq. (24) fixes the pixel mask along time because sensors do not move during a slot; reasonable for the model but an assumption.
  • ad hoc to paper Random pixel dropout approximates real sensor coverage
    Training masks individual pixels at random, while deployed sensors have contiguous sensing disks or grids; the mismatch is not tested.
  • ad hoc to paper Random patch masking is applied at inference
    Algorithm 1 randomly masks 75% of patches after pixel masking even at inference, discarding observed data without justification.
  • domain assumption Negligible Doppler at low user speed
    Section 3.2 states frequency shift is neglected for low-speed scenarios with user speeds of 1.0 to 1.5 m/s.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Temporal Spectrum Cartography in Low-Altitude Economy Networks: A Generative AI Framework with Multi-Agent Learning." pith.science (2026). https://pith.science/paper/2GNJ7ITK

@misc{pith2026250515571,
  author       = {Pith},
  title        = {Pith review of: Temporal Spectrum Cartography in Low-Altitude Economy Networks: A Generative AI Framework with Multi-Agent Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2GNJ7ITK}},
  note         = {Machine review of arXiv:2505.15571}
}
read the original abstract

This paper introduces a two-stage generative AI (GenAI) framework tailored for temporal spectrum cartography in low-altitude economy networks (LAENets). LAENets, characterized by diverse aerial devices such as UAVs, rely heavily on wireless communication technologies while facing challenges, including spectrum congestion and dynamic environmental interference. Traditional spectrum cartography methods have limitations in handling the temporal and spatial complexities inherent to these networks. Addressing these challenges, the proposed framework first employs a Reconstructive Masked Autoencoder (RecMAE) capable of accurately reconstructing spectrum maps from sparse and temporally varying sensor data using a novel dual-mask mechanism. This approach significantly enhances the precision of reconstructed radio frequency (RF) power maps. In the second stage, the Multi-agent Diffusion Policy (MADP) method integrates diffusion-based reinforcement learning to optimize the trajectories of dynamic UAV sensors. By leveraging temporal-attention encoding, this method effectively manages spatial exploration and exploitation to minimize cumulative reconstruction errors. Extensive numerical experiments validate that this integrated GenAI framework outperforms traditional interpolation methods and deep learning baselines by achieving 57.35% and 88.68% reconstruction error reduction, respectively. The proposed trajectory planner substantially improves spectrum map accuracy, reconstruction stability, and sensor deployment efficiency in dynamically evolving low-altitude environments.

Figures

Figures reproduced from arXiv: 2505.15571 by the authors.

Figure 1
Figure 1. System model of temporal spectrum cartography [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. The workflow of the proposed two-stage GenAI framework. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Overview of the RecMAE framework. The model processes masked spatiotemporal radio maps using 3D [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (7 more)
Figure 3
Figure 3. Figure 3: 4.1.6 Dual-mask Mechanism To robustly reconstruct radio maps from limited sensor measurements, we propose a dual-masking strategy that conceals information at two distinct scales: the patch (token) level and the pixel level. The pixel-level masking simulates limited se…
Figure 4
Figure 4. Figure 4: Overview of the MADP framework. The framework [PITH_FULL_IMAGE:figures/full_fig_p009_4.png]
Figure 5
Figure 5. Figure 5: The reconstruction comparison of different methods within 10-time slots and sensing radio [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 7
Figure 7. Figure 7: The training curve and cumulative MSE of different [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 6
Figure 6. Figure 6: The training curve and cumulative MSE of 4 methods [PITH_FULL_IMAGE:figures/full_fig_p013_6.png]
Figure 8
Figure 8. Figure 8: The training curve and cumulative MSE of different [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 9
Figure 9. Figure 9: The illustration of the MADP execution process and a UAV trajectory example. [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

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

  1. Vision-Aided ISAC in Low-Altitude Economy Networks via De-Diffused Visual Priors

    eess.SY 2025-07 reject novelty 6.0 of 10

    A DDPG-based resource scheduler for UAV ISAC networks uses de-diffused semantic tokens from images, fused with radar data, to select RAT, power, and beams, reaching within 4% of a raw-image oracle in simulation.

  2. Large Language Models for Next-Generation Wireless Network Management: A Survey and Tutorial

    cs.NI 2025-09 conditional novelty 4.0 of 10

    A survey and tutorial that organizes LLM-enabled wireless network optimization into formulation, solution, and verification stages, with case studies drawn from the authors' own prior papers.

  3. Toward Edge General Intelligence with Agentic AI and Agentification: Concepts, Technologies, and Future Directions

    cs.NI 2025-08 conditional novelty 4.0 of 10

    A survey that organizes agentic AI for 6G edge networks into four pillars, compactness, efficiency, knowledge and reasoning, and migration, and illustrates them with prior case studies.

  4. Toward Edge General Intelligence with Multiple-Large Language Model (Multi-LLM): Architecture, Trust, and Orchestration

    cs.NI 2025-07 conditional novelty 4.0 of 10

    A survey of multi-LLM systems in edge computing, covering architectures, enabling technologies, trust mechanisms, applications, and open datasets for edge general intelligence.

  5. Energy-Efficient RSMA-enabled Low-altitude MEC Optimization Via Generative AI-enhanced Deep Reinforcement Learning

    cs.NI 2025-07 conditional novelty 3.0 of 10

    A diffusion-enhanced SAC agent jointly optimizes UAV trajectory, offloading, and RSMA power allocation, reporting higher energy efficiency than NOMA/FDMA and DRL baselines in a simulated low-altitude MEC system.

Reference graph

Works this paper leans on

45 extracted references · 36 canonical work pages · cited by 5 Pith papers

  1. [2]

    Dynamic spectrum cartography: Reconstructing spatial-spectral-temporal radio frequency map via tensor completion,

    X. Chen, J. Wang, and Q. Huang, “Dynamic spectrum cartography: Reconstructing spatial-spectral-temporal radio frequency map via tensor completion,” IEEE Transactions on Signal Processing, 2025

    work page 2025

  2. [9]

    Spectrum surveying: Active radio map estimation with autonomous UAVs,

    R. Shrestha, D. Romero, and S. P . Chepuri, “Spectrum surveying: Active radio map estimation with autonomous UAVs,”IEEE trans- actions on wireless communications, vol. 22, no. 1, pp. 627–641, 2022

    work page 2022

  3. [28]

    UAV tra- jectory optimization for spectrum cartography: a PPO approach,

    L. Li, W. Li, J. Wang, X. Chen, Q. Peng, and W. Huang, “UAV tra- jectory optimization for spectrum cartography: a PPO approach,” IEEE Communications Letters, vol. 27, no. 6, pp. 1575–1579, 2023

    work page 2023

  4. [1]

    UAV swarm-enabled collaborative post-disaster com- munications in low altitude economy via a two-stage optimization approach,

    X. Zheng, G. Sun, J. Li, J. Wang, Q. Wu, D. Niyato, and A. Ja- malipour, “UAV swarm-enabled collaborative post-disaster com- munications in low altitude economy via a two-stage optimization approach,” arXiv preprint arXiv:2501.05742, 2025

    arXiv 2025

  5. [3]

    Efficient al- gorithms for air-to-ground channel reconstruction in UAV-aided communications,

    J. Chen, O. Esrafilian, D. Gesbert, and U. Mitra, “Efficient al- gorithms for air-to-ground channel reconstruction in UAV-aided communications,” in Proceedings of the 2017 IEEE Globecom Work- shops. IEEE, 2017, pp. 1–6

    work page 2017

  6. [4]

    Wireless sensor network for spectrum cartography based on krig- ing interpolation,

    G. Boccolini, G. Hernandez-Penaloza, and B. Beferull-Lozano, “Wireless sensor network for spectrum cartography based on krig- ing interpolation,” in Proceedings of the 2012 IEEE 23rd International Symposium on Personal, Indoor and Mobile Radio Communications . IEEE, 2012, pp. 1565–1570

    work page 2012

  7. [5]

    A comparison of in- terference cartography generation techniques in cognitive radio networks,

    S. ¨Ureten, A. Yongac ¸o˘glu, and E. Petriu, “A comparison of in- terference cartography generation techniques in cognitive radio networks,” in Proceedings of the 2012 IEEE International Conference on Communications. IEEE, 2012, pp. 1879–1883

    work page 2012

  8. [6]

    Deep spectrum cartography: Completing radio map tensors using learned neural models,

    S. Shrestha, X. Fu, and M. Hong, “Deep spectrum cartography: Completing radio map tensors using learned neural models,”IEEE Transactions on Signal Processing, vol. 70, pp. 1170–1184, 2022

    2022

Show all 45 references
  1. [7]

    Deep completion autoencoders for radio map estimation,

    Y. Teganya and D. Romero, “Deep completion autoencoders for radio map estimation,” IEEE Transactions on Wireless Communica- tions, vol. 21, no. 3, pp. 1710–1724, 2021

    2021

  2. [8]

    Spectrum cartography techniques, challenges, opportunities, and applications: A survey,

    Y. S. Reddy, A. Kumar, O. J. Pandey, and L. R. Cenkeramaddi, “Spectrum cartography techniques, challenges, opportunities, and applications: A survey,” Pervasive and Mobile Computing, vol. 79, p. 101511, 2022

    2022

  3. [10]

    Generative AI-enabled wireless communica- tions for robust low-altitude economy networking,

    C. Zhao, J. Wang, R. Zhang, D. Niyato, G. Sun, H. Du, D. I. Kim, and A. Jamalipour, “Generative AI-enabled wireless communica- tions for robust low-altitude economy networking,” arXiv preprint arXiv:2502.18118, 2025

    2025 arXiv

  4. [11]

    Videomae: Masked autoencoders are data-efficient learners for self-supervised video pre-training,

    Z. Tong, Y. Song, J. Wang, and L. Wang, “Videomae: Masked autoencoders are data-efficient learners for self-supervised video pre-training,” Advances in neural information processing systems , vol. 35, pp. 10 078–10 093, 2022

    2022

  5. [12]

    Efficient MAE towards large-scale vision transformers,

    Q. Han, G. Zhang, J. Huang, P . Gao, Z. Wei, and S. Lu, “Efficient MAE towards large-scale vision transformers,” in Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , 2024, pp. 606–615

    2024

  6. [13]

    Enhancing deep reinforcement learning: A tutorial on generative diffusion models in network optimization,

    H. Du, R. Zhang, Y. Liu, J. Wang, Y. Lin, Z. Li, D. Niyato, J. Kang, Z. Xiong, S. Cui et al., “Enhancing deep reinforcement learning: A tutorial on generative diffusion models in network optimization,” IEEE Communications Surveys & Tutorials, 2024

    2024

  7. [14]

    Masked autoencoders are scalable vision learners,

    K. He, X. Chen, S. Xie, Y. Li, P . Doll ´ar, and R. Girshick, “Masked autoencoders are scalable vision learners,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 16 000–16 009

    2022

  8. [15]

    Radio map estimation: A data-driven approach to spectrum cartography,

    D. Romero and S.-J. Kim, “Radio map estimation: A data-driven approach to spectrum cartography,” IEEE Signal Processing Maga- zine, vol. 39, no. 6, pp. 53–72, 2022

    2022

  9. [16]

    A joint tensor completion and prediction scheme for multi-dimensional spectrum map construction,

    M. Tang, G. Ding, Q. Wu, Z. Xue, and T. A. Tsiftsis, “A joint tensor completion and prediction scheme for multi-dimensional spectrum map construction,” IEEE Access , vol. 4, pp. 8044–8052, 2016

    2016

  10. [17]

    Tensor comple- tion for radio map reconstruction using low rank and smooth- ness,

    D. Sch ¨aufele, R. L. Cavalcante, and S. Stanczak, “Tensor comple- tion for radio map reconstruction using low rank and smooth- ness,” in Proceedings of the 2019 IEEE 20th International Workshop on 15 Signal Processing Advances in Wireless Communications. IEEE, 2019, pp. 1–5

    2019

  11. [18]

    Spectrum cartography via coupled block-term tensor decomposition,

    G. Zhang, X. Fu, J. Wang, X.-L. Zhao, and M. Hong, “Spectrum cartography via coupled block-term tensor decomposition,” IEEE Transactions on Signal Processing, vol. 68, pp. 3660–3675, 2020

    2020

  12. [19]

    A two-phase transfer learning- based power spectrum maps reconstruction algorithm for under- lay cognitive radio networks,

    X. Han, L. Xue, Y. Xu, and Z. Liu, “A two-phase transfer learning- based power spectrum maps reconstruction algorithm for under- lay cognitive radio networks,” IEEE Access , vol. 8, pp. 81 232– 81 245, 2020

    2020

  13. [20]

    RadioUNet: Fast radio map estimation with convolutional neural networks,

    R. Levie, C ¸ . Yapar, G. Kutyniok, and G. Caire, “RadioUNet: Fast radio map estimation with convolutional neural networks,” IEEE Transactions on Wireless Communications , vol. 20, no. 6, pp. 4001– 4015, 2021

    2021

  14. [21]

    Deep-learning-based radio map recon- struction for V2X communications,

    S. Roger, M. Brambilla, B. C. Tedeschini, C. Botella-Mascarell, M. Cobos, and M. Nicoli, “Deep-learning-based radio map recon- struction for V2X communications,” IEEE Transactions on Vehicular Technology, vol. 73, no. 3, pp. 3863–3871, 2023

    2023

  15. [22]

    Generative AI for secure physical layer com- munications: A survey,

    C. Zhao, H. Du, D. Niyato, J. Kang, Z. Xiong, D. I. Kim, X. Shen, and K. B. Letaief, “Generative AI for secure physical layer com- munications: A survey,” IEEE Transactions on Cognitive Communi- cations and Networking, 2024

    2024

  16. [23]

    A survey on vision transformer,

    K. Han, Y. Wang, H. Chen, X. Chen, J. Guo, Z. Liu, Y. Tang, A. Xiao, C. Xu, Y. Xu et al. , “A survey on vision transformer,” IEEE transactions on pattern analysis and machine intelligence, vol. 45, no. 1, pp. 87–110, 2022

    2022

  17. [24]

    Large language model (LLM) for telecommunications: A comprehensive survey on principles, key techniques, and opportunities,

    H. Zhou, C. Hu, Y. Yuan, Y. Cui, Y. Jin, C. Chen, H. Wu, D. Yuan, L. Jiang, D. Wu et al. , “Large language model (LLM) for telecommunications: A comprehensive survey on principles, key techniques, and opportunities,” IEEE Communications Surveys & Tutorials, 2024

    2024

  18. [25]

    Joint indoor localization and radio map construction with limited deployment load,

    S. Sorour, Y. Lostanlen, S. Valaee, and K. Majeed, “Joint indoor localization and radio map construction with limited deployment load,” IEEE Transactions on Mobile Computing , vol. 14, no. 5, pp. 1031–1043, 2014

    2014

  19. [26]

    3d radio map reconstruction and trajectory optimization for cellular-connected UAVs,

    Q. Gong, F. Wu, D. Yang, L. Xiao, and Z. Liu, “3d radio map reconstruction and trajectory optimization for cellular-connected UAVs,” Journal of Communications and Information Networks , vol. 8, no. 4, pp. 357–368, 2023

    2023

  20. [27]

    Online trajectory optimization for energy-efficient cellular-connected UAVs with map reconstruction,

    H. Zhao, Q. Hao, H. Huang, G. Gui, T. Ohtsuki, H. Sari, and F. Adachi, “Online trajectory optimization for energy-efficient cellular-connected UAVs with map reconstruction,” IEEE Trans- actions on Vehicular Technology, vol. 73, no. 3, pp. 3445–3456, 2023

    2023

  21. [29]

    Diffusion policy: Visuomotor policy learning via action diffusion,

    C. Chi, Z. Xu, S. Feng, E. Cousineau, Y. Du, B. Burchfiel, R. Tedrake, and S. Song, “Diffusion policy: Visuomotor policy learning via action diffusion,” The International Journal of Robotics Research, p. 02783649241273668, 2023

    2023

  22. [30]

    Generative AI enabled robust sensor placement in cyber-physical power systems: A graph diffusion approach,

    C. Zhao, G. Liu, B. Xiang, D. Niyato, B. Delinchant, H. Du, and D. I. Kim, “Generative AI enabled robust sensor placement in cyber-physical power systems: A graph diffusion approach,”arXiv preprint arXiv:2501.06756, 2025

    2025 arXiv

  23. [31]

    Enhancing physical layer communication security through generative AI with mixture of experts,

    C. Zhao, H. Du, D. Niyato, J. Kang, Z. Xiong, D. I. Kim, X. S. Shen, and K. B. Letaief, “Enhancing physical layer communication security through generative AI with mixture of experts,” IEEE Wireless Communications, 2025

    2025

  24. [33]

    Generative AI for integrated sensing and communica- tion: Insights from the physical layer perspective,

    J. Wang, H. Du, D. Niyato, J. Kang, S. Cui, X. S. Shen, and P . Zhang, “Generative AI for integrated sensing and communica- tion: Insights from the physical layer perspective,” IEEE Wireless Communications, 2024

    2024

  25. [34]

    Real-time outdoor localization using radio maps: A deep learning approach,

    C ¸ . Yapar, R. Levie, G. Kutyniok, and G. Caire, “Real-time outdoor localization using radio maps: A deep learning approach,” IEEE Transactions on Wireless Communications, vol. 22, no. 12, pp. 9703– 9717, 2023

    2023

  26. [35]

    A new algorithm for indoor RSSI radio map reconstruction,

    W. Xue, Q. Li, X. Hua, K. Yu, W. Qiu, and B. Zhou, “A new algorithm for indoor RSSI radio map reconstruction,” IEEE Access, vol. 6, pp. 76 118–76 125, 2018

    2018

  27. [36]

    3GPP TR 38.901 channel model,

    Q. Zhu, C.-X. Wang, B. Hua, K. Mao, S. Jiang, and M. Yao, “3GPP TR 38.901 channel model,” in the wiley 5G Ref: the essential 5G reference online. Wiley Press, 2021, pp. 1–35

    2021

  28. [37]

    Correlation model for shadow fading in mo- bile radio systems,

    M. Gudmundson, “Correlation model for shadow fading in mo- bile radio systems,” Electronics letters , vol. 27, no. 23, pp. 2145– 2146, 1991

    1991

  29. [38]

    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,” Advances in neural information processing systems , vol. 30, 2017

    2017

  30. [39]

    Analyzing multi-head self-attention: Specialized heads do the heavy lifting, the rest can be pruned,

    E. Voita, D. Talbot, F. Moiseev, R. Sennrich, and I. Titov, “Analyzing multi-head self-attention: Specialized heads do the heavy lifting, the rest can be pruned,” arXiv preprint arXiv:1905.09418, 2019

    1905 arXiv

  31. [40]

    Partially observable markov decision processes,

    M. T. Spaan, “Partially observable markov decision processes,” in Reinforcement learning: State-of-the-art. Springer, 2012, pp. 387–414

    2012

  32. [41]

    Denoising diffusion probabilistic models,

    J. Ho, A. Jain, and P . Abbeel, “Denoising diffusion probabilistic models,” Advances in neural information processing systems , vol. 33, pp. 6840–6851, 2020

    2020

  33. [42]

    Multi-agent actor-critic for mixed cooperative-competitive environments,

    R. Lowe, Y. I. Wu, A. Tamar, J. Harb, O. Pieter Abbeel, and I. Mor- datch, “Multi-agent actor-critic for mixed cooperative-competitive environments,” Advances in neural information processing systems , vol. 30, 2017

    2017

  34. [43]

    Deterministic policy gradient algorithms,

    D. Silver, G. Lever, N. Heess, T. Degris, D. Wierstra, and M. Ried- miller, “Deterministic policy gradient algorithms,” in International conference on machine learning. Pmlr, 2014, pp. 387–395

    2014

  35. [44]

    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, Technical Report TR 38.901, Nov. 2020, release 16

    2020

  36. [45]

    High-efficiency urban 3D radio map estimation based on sparse measurements,

    X. Chen, X. Zhong, Z. Zhang, L. Dai, and S. Zhou, “High-efficiency urban 3D radio map estimation based on sparse measurements,” arXiv preprint arXiv:2408.04205, 2024

    2024 arXiv

  37. [46]

    3-D placement of an unmanned aerial vehicle base station for maximum coverage of users with different QoS requirements,

    M. Alzenad, A. El-Keyi, and H. Yanikomeroglu, “3-D placement of an unmanned aerial vehicle base station for maximum coverage of users with different QoS requirements,” IEEE Wireless Commu- nications Letters, vol. 7, no. 1, pp. 38–41, 2017

    2017

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.