Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Hybrid Near-Far Field 6D Movable Antenna Design Exploiting Directional Sparsity and Deep Learning

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a hybrid-field channel model, which treats each movable surface as far-field and the gaps between surfaces as near-field, lets six-dimensional movable antenna arrays approach near-field sum rates at far-field…

desk verdict The hybrid-field 6DMA model and directional-sparsity channel estimator are worth engaging; the A2C algorithm as written cannot sample its own continuous action space, so the DRL results need a described fix before they are reproducible. read the letter →

arxiv 2506.15808 v1 pith:KXFXS25S submitted 2025-06-18 cs.IT eess.SPmath.IT

Xiaodan Shao , Limei Hu , Yulong Sun , Xing Li , Yixiao Zhang , Jingze Ding , Xiaoming Shi , Feng Chen
show 2 more authors
Derrick Wing Kwan Ng Robert Schober
This is my paper · ORCID
classification cs.ITeess.SPmath.IT MSC 94A4068T0790C4094A12
keywords 6Dmovableantennahybrid-fieldchannelmodelnear-fieldcommunicationdirectionalsparsityestimationdeepreinforcementlearningA2Csum-rateoptimization
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

The paper is trying to establish that six-dimensional movable antenna (6DMA) systems can be made practical by refusing to choose between far-field and near-field channel models. It proposes a hybrid-field model in which each movable surface keeps a planar-wave steering vector while the propagation between surfaces is spherical, cutting the number of channel parameters relative to the full near-field model. On top of this model, it builds a channel-estimation algorithm that exploits directional sparsity to reconstruct the channel at unmeasured surface configurations, and a deep-reinforcement-learning controller that jointly chooses surface positions, rotations, and beamforming. If the claims hold, 6DMA base stations could get near-field-level sum rates with far-field-level modeling and optimization cost.

What carries the argument

The working object is the hybrid-field channel vector $\mathbf{h}(q,u)$ of equation (13), which combines per-surface planar-wave steering vectors $\mathbf{a}(q_b,u_b)$ with per-surface spherical-wave distances $d_b$ and directional antenna gains $g(q_b,u_b)$. Directional sparsity is the second mechanism: the antenna gain is treated as exactly zero for position-rotation pairs whose surface does not face the user, which turns the channel-estimation problem into one of localizing the user from a few strong measurements. The estimation pipeline is carried by a coarse-grid maximum-likelihood search per surface, a distance-based clustering step that selects the largest consistent cluster, and a fine-grid joint search over that cluster. The optimization is carried by an A2C agent with an actor network for actions, a critic network for state values, and reward penalties for violating position, spacing, and rotation constraints.

What would settle it

Measure the channel from a single user to many candidate 6DMA position-rotation pairs in an environment with one strong reflector, using the same antenna pattern as in the simulations, and compare each surface's received power against the zero-gain prediction; if surfaces whose main beam points away from the user still receive power well above the noise floor, directional sparsity fails and Algorithm 1's cluster selection and channel-map reconstruction would be biased.

Watch

Extended reading notes

Core claim

The central claim is that a user in a 6DMA system can simultaneously be in the far field of any one movable surface and in the near field of the collection of surfaces, so the right channel model is hybrid. The proposed model (13) assigns each surface one azimuth-elevation pair $(\phi_b, \theta_b)$, one distance $d_b$ for the spherical phase, and a planar-wave steering vector $\mathbf{a}(q_b,u_b)$ built from the surface's position and rotation, which preserves spatial separation between surfaces without modeling every antenna pair as spherical. The paper also claims that 6DMA channels are directionally sparse: only surfaces whose antenna gain toward the user is nonzero carry significant power, which lets the estimator discard uninformative position-rotation pairs. Its estimation algorithm first obtains maximum-likelihood user-location estimates surface by surface, clusters those estimates to reject outliers, and refines the parameters using only the largest cluster, then reconstructs the full channel map for all candidate configurations. Finally, the paper claims that an Advantage Actor-Critic deep-reinforcement-learning agent with penalty-shaped rewards can jointly optimize positions, rotations, and beamforming, and simulations show the hybrid-field model approaching near-field sum rates while clearly outperforming fixed, fluid, and rotation-only antenna systems.

Load-bearing premise

The load-bearing premise is that a 6DMA surface whose antenna pattern does not point at the user has exactly zero channel gain, so its measurements can be discarded; if real antennas leak power through side lobes or multipath reflects energy into misaligned surfaces, the clustering step may discard useful information and bias the reconstructed channel map.

Editorial extensions

If this is right

  • The hybrid-field model (13) achieves sum rates close to the full near-field model while using fewer channel parameters, so 6DMA design can use the cheaper model without losing near-field performance.
  • Channel information at all candidate position-rotation pairs can be reconstructed from measurements at only $M$ configurations, cutting the training overhead caused by physically moving the surfaces.
  • Directional-sparsity clustering rejects outlier surface estimates and gives lower channel-estimation MSE than least-squares estimation, so the method scales to many candidate configurations.
  • The A2C-based joint design outperforms fixed-position, fluid-antenna, and rotation-only systems, with the gain widening in the near field, at larger base-station site sizes, and with more users.
  • Unlike fluid antenna systems, whose sum rate plateaus when the site grows, the proposed 6DMA system keeps improving with site size, indicating continued use of extra spatial degrees of freedom.

Reading between the lines

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

  • The paper leaves implicit that the hybrid-field modeling principle is a general recipe: any large-aperture array built from movable subarrays could keep plane-wave steering inside each subarray and spherical-wave phases between subarrays, which would extend to movable arrays on drones or other deployable platforms.
  • A stress test the paper does not run is rich multipath, where directional sparsity weakens; in such environments an estimator that retains weak-gain surfaces with small weights instead of discarding them might be more robust.
  • The paper does not report whether the learned policy transfers across user hotspot geometries, even though the A2C state includes user locations; a transfer experiment could quantify how much retraining is needed when the spatial distribution of users changes.
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

3 major / 5 minor

Summary. The paper proposes a hybrid near-far field channel model for six-dimensional movable antenna (6DMA) systems, where users may be in the far field of individual 6DMA surfaces but in the near field of the overall array. It also proposes a low-overhead channel estimation algorithm that exploits directional sparsity to reconstruct the channel map over all candidate position-rotation pairs, and an A2C-based deep reinforcement learning algorithm to jointly optimize the 6DMA positions, rotations, and transmit beamforming. Numerical results are presented to support the claims that the hybrid-field model approaches the near-field model's sum rate with fewer parameters, that the estimation algorithm outperforms LS-based baselines, and that the DRL-enhanced 6DMA system outperforms flexible antenna systems.

Significance. If the claims hold, the hybrid-field model offers a practical complexity-accuracy trade-off for 6DMA channel modeling, and the directional-sparsity-based estimation is an interesting low-overhead approach for a continuously movable antenna array. The paper also identifies a relevant operational regime where users are in the near field of some surfaces but the far field of others. However, the central DRL contribution is not reproducible as written because the actor network's softmax output is incompatible with the continuous action space, and the channel estimation relies on an exact zero-gain sparsity assumption that is not reflected in the 3GPP antenna pattern used in simulations. These gaps prevent verification of the main performance claims.

major comments (3)
  1. [IV-B3, IV-B4, Algorithm 2] The action space defined in Eq. (37) is continuous, containing 3B position entries, 3B rotation entries, and 2K(NB) real beamforming entries. The actor network, however, is described in Section IV-B4 as using a softmax function to output action probabilities, from which actions are sampled. A softmax layer defines a probability distribution over a finite set of categories, not over a continuous Euclidean space. No discretization grid, continuous reparameterization, or Gaussian/other continuous policy is specified. Consequently, the policy π_{z_a}(a_t|s_t) used in Eqs. (38)-(42) is undefined for almost every action, and line 9 of Algorithm 2 cannot be executed. The normalization layer mentioned in Section IV-B3 only enforces the sum-power constraint on the beamforming part and does not resolve how continuous q and u are sampled, nor how constraints (31c)-(31e) are enforced during sampling. This is an internal inconsistency in the description of the algorithm that the simulation results in Figs. 6-11 depend on. The authors should specify the exact policy parameterization (e.g., a Gaussian policy with mean and variance outputs, or an explicit discretization of the action space with a concrete grid and the resulting finite action set), and state how the position and rotation constraints are imposed when sampling.
  2. [III-A and Section V] The channel estimation algorithm relies on directional sparsity, where the antenna gain g(q_m,u_m) is taken to be exactly zero for 6DMA position-rotation pairs not in the set W, i.e., for surfaces not oriented toward the user. However, the simulation setup uses the 3GPP antenna pattern [50], which has nonzero sidelobes in all directions, and the paper does not quantify how much the zero-gain approximation deviates from the actual pattern. The clustering step in Algorithm 1 explicitly discards estimates that fall outside the dominant cluster; if weak but nonzero-gain surfaces produce estimates that are not colocated with the true user position, this outlier rejection could bias the reconstructed channel map. Since the zero-gain assumption is load-bearing for the claimed estimation performance in Fig. 5 and for the subsequent DRL design, the authors should justify the approximation for the 3GPP pattern, or provide a robustness study (e.g., varying the threshold ε and showing the sensitivity of MSE and sum rate to the sidelobe level).
  3. [V, Fig. 8] Fig. 8 is used to support the claim that the hybrid-field model 'closely approaches' the near-field model. The evaluation applies the proposed DRL method to three channel models and then evaluates the resulting positions, rotations, and beamformers under the near-field channel. Because the DRL algorithm itself is not implementable as written (see the first major comment), the optimized configurations used in Fig. 8 are not reproducible, and the comparison among the three models is not verifiable. Even setting that aside, the simulation currently does not report the number of channel parameters used by each model, so the claim of 'fewer parameters' is not directly evidenced in the figure; a parameter-count comparison would strengthen the argument.
minor comments (5)
  1. [IV-C] Section IV-C states that 'We outline detailed pseudocode of the proposed DRL algorithm's in Algorithm 1', but the DRL pseudocode is actually Algorithm 2; Algorithm 1 is the channel estimation procedure. The reference should be corrected.
  2. [II-C, Eq. (18)] Equation (18) has a parenthesis mismatch in the numerator: it reads '|(h_k(q,u)^H w_k|^2' instead of '|h_k(q,u)^H w_k|^2'. This should be corrected for readability.
  3. [IV-B2, Eq. (36)] The state space decomposition for transmission power is described by splitting ∥w_k∥^2 into real and imaginary parts, but w_k^H w_k is a real scalar, so its imaginary part is identically zero. The description of the state entries (2K entries for 'transmission power' plus 2K^2 for received power) should be revised to reflect that the actual state dimension is 3K + K (for the real-valued power terms) plus 3K for locations, rather than 5K + 2K^2 with the stated decomposition.
  4. [V, Table I/II] The text refers to 'Table II lists the key simulation parameters', but the manuscript contains only one table, labeled Table I. The table numbering and the in-text reference should be made consistent.
  5. [V, Fig. 4 caption] The caption of Fig. 4 appears incomplete: it reads 'Channel gain (dB)' and the axis label 'M N' is unclear. The caption should describe what is plotted and the meaning of the axes.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the hybrid-field model, channel estimator, and DRL design are self-contained constructions validated against external benchmarks.

full rationale

The paper's central derivation chain is not circular. The hybrid-field channel model in Eq. (13) is an explicit geometric construction that combines a per-surface planar-wave steering vector with per-surface spherical-wave path phases and gains; it is not fitted to the near-field model and, as Remark 1 states, it reduces to the far-field model when B=1 and to the near-field model when N=1. The channel estimation algorithm in Section III is standard maximum-likelihood parameter estimation: Eq. (22) searches a grid for (d, phi, theta), Eq. (21) gives the closed-form path gain, and Eq. (30) reconstructs the channel by inserting those estimates into the same hybrid-field model. This is parameter estimation under the assumed model, not a prediction that is equivalent to its own input. The directional-sparsity assumption g(q_m,u_m)=0 for non-visible surfaces is an explicit modeling approximation drawn from prior 6DMA work and is testable against the 3GPP antenna patterns used in simulation; it is not a self-referential definition of the target quantity. The DRL contribution is a standard A2C policy-gradient update in Eqs. (38)-(44), with the reward defined directly from the sum-rate objective and constraint penalties. The skeptical observation that the continuous action space in Eq. (37) is not obviously compatible with the softmax output described in Section IV-B4 is a real implementation-level inconsistency and a correctness risk, but it is not circularity: nothing in the derivation reduces to assuming the conclusion. Finally, the numerical validation uses external benchmarks: Fig. 8 evaluates configurations on the ground-truth near-field channel model, and Figs. 9-11 compare against FPA, FAS, and rotatable-6DMA baselines. Thus the paper's claims are not forced by construction, and no load-bearing self-citation chain is used to establish the central results.

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

The paper introduces no new physical entities. The main load-bearing assumptions are the LoS single-path channel model, the directional sparsity zero-gain approximation, and the validity of the hybrid-field model itself. The free parameters are algorithm hyperparameters and simulation settings chosen by hand, whose values are mostly undisclosed.

free parameters (6)
  • Coarse grid step sizes (Delta d_H, Delta phi_H, Delta theta_H) = not specified
    Used in (20) to define the coarse search grid for ML estimation. Chosen by hand; values are not reported.
  • Fine grid step sizes (Delta d_L, Delta phi_L, Delta theta_L) and ranges (D, Phi, Theta) = not specified
    Used in (28) for the refinement. Defaults are not reported.
  • Clustering threshold epsilon = not specified
    Determines cluster formation in Algorithm 1 and is required for outlier rejection. No value is given.
  • DRL learning rates mu_a, mu_c = 0.1, 0.01, 1e-5 depending on reward sign
    Hand-designed schedule in (45).
  • Number of candidate position-rotation pairs M = 16 or 25 in simulations
    Used for channel estimation training measurements.
  • Noise power sigma^2 = -80 dBm
    Simulation parameter that affects SNR and estimation accuracy.
assumptions (6)
  • domain assumption Line-of-sight single-path channel between each user and the BS
    Stated in Section II-B; no multipath or scattering is modeled. The channel models in (6), (10), and (13) all assume one propagation path.
  • domain assumption Directional sparsity: antenna gain g(q_m,u_m)=0 for surfaces not oriented toward the user
    Section III-A. This approximation is used to ignore weak surfaces in estimation. Real antenna patterns have sidelobes, so it is a simplification.
  • domain assumption The hybrid-field model is a valid approximation: planar wave within a surface, spherical among surfaces
    Eq. (13). This is the core modeling choice; its accuracy is only tested by simulation.
  • domain assumption The near-field model in (10) is the ground truth
    Used in Fig. 8 and later evaluations. It is also a LoS geometric model.
  • domain assumption BS region C is a cube of side 2 m and users lie in a spherical annulus 20-200 m
    Simulation setup in Section V; the Rayleigh distance is 192 m, so users are partly in the near-field.
  • domain assumption 6DMA surfaces can be freely positioned and rotated within the cube, subject to constraints (31c)-(31e)
    The DRL agent learns to satisfy these constraints via penalties in (32)-(35).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hybrid Near-Far Field 6D Movable Antenna Design Exploiting Directional Sparsity and Deep Learning." pith.science (2026). https://pith.science/paper/KXFXS25S

@misc{pith2026250615808,
  author       = {Pith},
  title        = {Pith review of: Hybrid Near-Far Field 6D Movable Antenna Design Exploiting Directional Sparsity and Deep Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KXFXS25S}},
  note         = {Machine review of arXiv:2506.15808}
}
read the original abstract

Six-dimensional movable antenna (6DMA) has been identified as a new disruptive technology for future wireless systems to support a large number of users with only a few antennas. However, the intricate relationships between the signal carrier wavelength and the transceiver region size lead to inaccuracies in traditional far-field 6DMA channel model, causing discrepancies between the model predictions and the hybrid-field channel characteristics in practical 6DMA systems, where users might be in the far-field region relative to the antennas on the same 6DMA surface, while simultaneously being in the near-field region relative to different 6DMA surfaces. Moreover, due to the high-dimensional channel and the coupled position and rotation constraints, the estimation of the 6DMA channel and the joint design of the 6DMA positions and rotations and the transmit beamforming at the base station (BS) incur extremely high computational complexity. To address these issues, we propose an efficient hybrid-field generalized 6DMA channel model, which accounts for planar-wave propagation within individual 6DMA surfaces and spherical-wave propagation among different 6DMA surfaces. Furthermore, by leveraging directional sparsity, we propose a low-overhead channel estimation algorithm that efficiently constructs a complete channel map for all potential antenna position-rotation pairs while limiting the training overhead incurred by antenna movement. In addition, we propose a low-complexity design leveraging deep reinforcement learning (DRL), which facilitates the joint design of the 6DMA positions, rotations, and beamforming in a unified manner. Numerical results demonstrate that the proposed hybrid-field channel model and channel estimation algorithm outperform existing approaches and that the DRL-enhanced 6DMA system significantly surpasses flexible antenna systems.

Figures

Figures reproduced from arXiv: 2506.15808 by the authors.

Figure 1
Figure 1. (a) A typical hybrid-field propagation environment in a 6DMA THz [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Deep reinforcement learning for joint optimization of 6DMAs’ positions, rotations, and beamforming. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Proposed MLP structure of the critic network and the actor network. [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (5 more)
Figure 5
Figure 5. Figure 5: The MSE of 6DMA channel estimation versus SNR. [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 8
Figure 8. Figure 8: Sum rate versus the number of 6DMA surfaces in the near-field 6DMA [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 7
Figure 7. Figure 7: Rewards versus time steps under different number of users. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 10
Figure 10. Figure 10: Sum rate versus the proportion of regular users. [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]
Figure 11
Figure 11. Figure 11: Sum rate versus the number of users. BS size [PITH_FULL_IMAGE:figures/full_fig_p012_11.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. An Efficient Model-Driven Groupwise Approach for Atlas Construction

    cs.CV 2025-08 reject novelty 5.0 of 10

    The abstract and full text are two different papers, so the claimed DARC atlas-construction framework is never presented or supported.

Reference graph

Works this paper leans on

50 extracted references · 30 canonical work pages · cited by 1 Pith paper

  1. [50]

    Study on 3D channel model for LTE,

    G. T. 36.873:, “Study on 3D channel model for LTE,” vol. V12.7.0, Dec. 2017

    work page 2017

  2. [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. Tutor., pp. 1–1, Jan. 2024

    work page 2024

  3. [2]

    Massive MIMO for next generation wireless systems,

    E. G. Larsson, O. Edfors, F. Tufvesson, and T. L. Marzetta, “Massive MIMO for next generation wireless systems,”IEEE Commun. Mag., vol. 52, no. 2, pp. 186–195, Feb. 2014

    2014

  4. [3]

    Next generation advanced transceiver technologies for 6G and beyond,

    C. Youet al., “Next generation advanced transceiver technologies for 6G and beyond,”IEEE J. Sel. Areas Commun., vol. 43, no. 3, pp. 582–627, Mar. 2025

    2025

  5. [4]

    Channel estimation for extremely large-scale massive MIMO systems,

    Y . Han, S. Jin, C.-K. Wen, and X. Ma, “Channel estimation for extremely large-scale massive MIMO systems,”IEEE Wirel. Commun. Lett., vol. 9, no. 5, pp. 633–637, May 2020

    work page 2020

  6. [5]

    Terahertz multi-user massive MIMO with intelligent reflecting surface: Beam training and hybrid beamforming,

    B. Ning, Z. Chen, W. Chen, Y . Du, and J. Fang, “Terahertz multi-user massive MIMO with intelligent reflecting surface: Beam training and hybrid beamforming,”IEEE Trans. Veh. Technol., vol. 70, no. 2, pp. 1376–1393, Feb. 2021

    work page 2021

  7. [6]

    Target sensing with intelligent reflecting surface: Architecture and performance,

    X. Shao, C. You, W. Ma, X. Chen, and R. Zhang, “Target sensing with intelligent reflecting surface: Architecture and performance,”IEEE J. Sel. Areas Commun., vol. 40, no. 7, pp. 2070–2084, Jul. 2022

    2022

  8. [7]

    Scalable cell-free massive MIMO systems,

    E. Bjornson and L. Sanguinetti, “Scalable cell-free massive MIMO systems,”IEEE Trans. Commun., vol. 68, no. 7, pp. 4247–4261, Jul. 2020

    work page 2020

Show all 50 references
  1. [8]

    Hybrid digital and analog beamforming design for large-scale mimo systems,

    F. Sohrabi and W. Yu, “Hybrid digital and analog beamforming design for large-scale mimo systems,” inIEEE Int. Conf. Acoust. Speech Signal Process. (ICASSP), Apr. 2015, pp. 2929–2933

    2015

  2. [9]

    Full-angle digital predistortion of 5G millimeter-wave massive MIMO transmitters,

    C. Yu, J. Jing, H. Shao, Z. H. Jiang, P. Yan, X.-W. Zhu, W. Hong, and A. Zhu, “Full-angle digital predistortion of 5G millimeter-wave massive MIMO transmitters,”IEEE Trans. Microw. Theory Tech., vol. 67, no. 7, pp. 2847–2860, Jul. 2019

    2019

  3. [10]

    6D movable antenna based on user distribution: Modeling and optimization,

    X. Shao, Q. Jiang, and R. Zhang, “6D movable antenna based on user distribution: Modeling and optimization,”IEEE Trans. Wireless Commun., vol. 24, no. 1, pp. 355–370, Jan. 2025

    2025

  4. [11]

    6D movable antenna enhanced wireless network via discrete position and rotation optimiza- tion,

    X. Shao, R. Zhang, Q. Jiang, and R. Schober, “6D movable antenna enhanced wireless network via discrete position and rotation optimiza- tion,”IEEE J. Sel. Areas Commun., vol. 43, no. 3, pp. 674–687, Mar. 2025

    2025

  5. [12]

    Exploiting six-dimensional mov- able antenna for wireless sensing,

    X. Shao, R. Zhang, and R. Schober, “Exploiting six-dimensional mov- able antenna for wireless sensing,”IEEE Wireless Commun. Lett., vol. 14, no. 2, pp. 265–269, Feb. 2025

    2025

  6. [13]

    6DMA enhanced wireless network with flexible antenna position and rotation: Opportunities and challenges,

    X. Shao and R. Zhang, “6DMA enhanced wireless network with flexible antenna position and rotation: Opportunities and challenges,” vol. 63, no. 4, pp. 121–128, Apr. 2025

    2025

  7. [14]

    Capacity maximization for base station with hybrid fixed and movable antennas,

    X. Shi, X. Shao, and R. Zhang, “Capacity maximization for base station with hybrid fixed and movable antennas,” vol. 13, no. 10, pp. 2877–2881, Oct. 2024

    2024

  8. [15]

    Distributed channel estimation for 6D movable antenna: Unveiling directional sparsity,

    X. Shao, R. Zhang, Q. Jiang, J. Park, Q. T. Q. S., and R. Schober, “Distributed channel estimation for 6D movable antenna: Unveiling directional sparsity,”arXiv preprint arXiv:2409.16510, 2024

    2024 arXiv

  9. [16]

    Polarized 6D movable antenna for wireless communication: Channel modeling and optimization,

    X. Shao, Q. Jiang, D. W. K. Ng, and N. Al-Dhahir, “Polarized 6D movable antenna for wireless communication: Channel modeling and optimization,”arXiv preprint arXiv:2506.04471, 2025. 13

    2025 arXiv

  10. [17]

    Polarforming antenna enhanced sensing and communication: Modeling and optimization,

    X. Shao, R. Zhang, H. Zhou, Q. Jiang, C. Zhou, W. Zhuang, and X. Shen, “Polarforming antenna enhanced sensing and communication: Modeling and optimization,”arXiv preprint arXiv:2505.08070, 2025

    2025 arXiv

  11. [18]

    UA V-enabled passive 6D movable antennas: Joint deployment and beamforming optimization,

    C. Liu, W. Mei, P. Wang, Y . Meng, B. Ning, and Z. Chen, “UA V-enabled passive 6D movable antennas: Joint deployment and beamforming optimization,”arXiv preprint arXiv:2412.11150, 2024

    2024 arXiv

  12. [19]

    Intelligent reflecting surface-aided wireless communication with movable elements,

    G. Hu, Q. Wu, D. Xuet al., “Intelligent reflecting surface-aided wireless communication with movable elements,”IEEE Wireless Commun. Lett., vol. 13, no. 4, pp. 1173–1177, Apr. 2024

    2024

  13. [20]

    Wireless communi- cation with flexible reflector: Joint placement and rotation optimization for coverage enhancement,

    H. Lu, Z. Yu, Y . Zeng, S. Ma, S. Jin, and R. Zhang, “Wireless communi- cation with flexible reflector: Joint placement and rotation optimization for coverage enhancement,”arXiv preprint arXiv:2412.18817, 2024

    2024 arXiv

  14. [21]

    A tutorial on six-dimensional mov- able antenna for 6G networks: Synergizing positionable and rotatable antennas,

    X. Shao, W. Mei, C. Youet al., “A tutorial on six-dimensional mov- able antenna for 6G networks: Synergizing positionable and rotatable antennas,”arXiv preprint arXiv:2503.18240, 2025

    2025 arXiv

  15. [22]

    UA V-enabled passive 6DMA for ISAC: Joint location, orientation, and reflection optimization,

    P. Wang, Y . Xue, W. Mei, J. Fang, and R. Zhang, “UA V-enabled passive 6DMA for ISAC: Joint location, orientation, and reflection optimization,”arXiv preprint arXiv:2505.10220, 2025

    2025

  16. [23]

    6DMA-aided cell-free massive MIMO communication,

    X. Shiet al., “6DMA-aided cell-free massive MIMO communication,” early access, IEEE Wireless Commun. Lett., pp. 1–1, 2025

    2025

  17. [24]

    Passive six-dimensional movable antenna (6dma)- assisted multiuser communication,

    H. Wanget al., “Passive six-dimensional movable antenna (6dma)- assisted multiuser communication,”IEEE Wireless Commun. Lett., vol. 14, no. 4, pp. 1014–1018, Apr. 2025

    2025

  18. [25]

    Wireless communi- cation with flexible reflector: Joint placement and rotation optimization for coverage enhancement,

    H. Lu, Z. Yu, Y . Zeng, S. Ma, S. Jin, and R. Zhang, “Wireless communi- cation with flexible reflector: Joint placement and rotation optimization for coverage enhancement,”IEEE Trans. Wireless Commun., pp. 1–1, 2025

    2025

  19. [26]

    6DMA-aided hybrid beamforming with joint antenna position and orientation optimization,

    Y . Zhang, Y . Zhang, L. Zhu, S. Xiao, W. Tang, Y . C. Eldar, and R. Zhang, “6DMA-aided hybrid beamforming with joint antenna position and orientation optimization,”arXiv preprint arXiv:2412.17088, 2024

    2024 arXiv

  20. [27]

    A tutorial on movable antennas for wireless networks,

    L. Zhuet al., “A tutorial on movable antennas for wireless networks,” IEEE Commun. Surv. Tutor., pp. 1–1, 2025

    2025

  21. [28]

    Performance analysis and optimization for movable antenna aided wideband communications,

    L. Zhu, W. Ma, Z. Xiao, and R. Zhang, “Performance analysis and optimization for movable antenna aided wideband communications,” IEEE Trans. Wireless Commun., vol. 23, no. 12, pp. 18 653–18 668, Dec. 2024

    2024

  22. [29]

    Globally optimal movable antenna-enhanced multi-user communication: Discrete antenna positioning, motion power consumption, and imperfect CSI,

    Y . Wu, D. Xu, D. W. K. Ng, W. Gerstacker, and R. Schober, “Globally optimal movable antenna-enhanced multi-user communication: Discrete antenna positioning, motion power consumption, and imperfect CSI,” arXiv preprint arXiv:2408.15435, 2024

    2024 arXiv

  23. [30]

    Capacity maximization for FAS-assisted multiple access channels,

    H. Xuet al., “Capacity maximization for FAS-assisted multiple access channels,”IEEE Trans. Commun., pp. 1–1, 2024

    2024

  24. [31]

    Secure wireless communication via movable-antenna array,

    G. Hu, Q. Wu, K. Xu, J. Si, and N. Al-Dhahir, “Secure wireless communication via movable-antenna array,”IEEE Signal Process. Lett., vol. 31, pp. 516–520, Jan. 2024

    2024

  25. [32]

    Movable-antenna position optimization for physical-layer security via discrete sampling,

    W. Mei, X. Wei, Y . Liu, B. Ning, and Z. Chen, “Movable-antenna position optimization for physical-layer security via discrete sampling,” arXiv preprint arXiv:2408.08322, 2024

    2024 arXiv

  26. [33]

    Movable antenna enabled integrated sensing and communication,

    W. Lyu, S. Yang, Y . Xiu, Z. Zhang, C. Assi, and C. Yuen, “Movable antenna enabled integrated sensing and communication,”IEEE Trans. Wireless Commun., vol. 24, no. 4, pp. 2862–2875, Apr. 2025

    2025

  27. [34]

    Channel estimation for extremely large-scale massive MIMO: Far-field, near-field, or hybrid-field?

    X. Wei and L. Dai, “Channel estimation for extremely large-scale massive MIMO: Far-field, near-field, or hybrid-field?”IEEE Commun. Lett., vol. 26, no. 1, pp. 177–181, Jan. 2022

    2022

  28. [35]

    Beam focusing for near-field multiuser mimo com- munications,

    H. Zhanget al., “Beam focusing for near-field multiuser mimo com- munications,”IEEE Trans. Wireless Commun., vol. 21, no. 9, pp. 7476– 7490, Sept. 2022

    2022

  29. [36]

    Channel estimation for extremely large-scale MIMO: Far-field or near-field?

    M. Cui and L. Dai, “Channel estimation for extremely large-scale MIMO: Far-field or near-field?”IEEE Trans. Commun., vol. 70, no. 4, pp. 2663–2677, Apr. 2022

    2022

  30. [37]

    Near-field MIMO communications for 6G: Fundamentals, challenges, potentials, and future directions,

    M. Cui, Z. Wu, Y . Lu, X. Wei, and L. Dai, “Near-field MIMO communications for 6G: Fundamentals, challenges, potentials, and future directions,”IEEE Commun. Maga., vol. 61, no. 1, pp. 40–46, Jan. 2023

    2023

  31. [38]

    Near- field communications: A tutorial review,

    Y . Liu, Z. Wang, J. Xu, C. Ouyang, X. Mu, and R. Schober, “Near- field communications: A tutorial review,”IEEE Open J. Commun. Soc., vol. 4, pp. 1999–2049, Aug. 2023

    1999

  32. [39]

    Feature-aided adaptive-tuning deep learning for massive device detection,

    X. Shao, X. Chen, Y . Qiang, C. Zhong, and Z. Zhang, “Feature-aided adaptive-tuning deep learning for massive device detection,”IEEE J. Sel. Areas Commun., vol. 39, no. 7, pp. 1899–1914, Jul. 2021

    1914

  33. [40]

    Intelligent surfaces empowered wireless network: Recent advances and the road to 6G,

    Q. Wuet al., “Intelligent surfaces empowered wireless network: Recent advances and the road to 6G,”Proc. IEEE, vol. 112, no. 7, pp. 724–763, Jul. 2024

    2024

  34. [41]

    Reinforcement learning meets wireless networks: A layering perspective,

    Y . Chen, Y . Liu, M. Zeng, U. Saleem, Z. Lu, X. Wen, D. Jin, Z. Han, T. Jiang, and Y . Li, “Reinforcement learning meets wireless networks: A layering perspective,”IEEE Int. Things J., vol. 8, no. 1, pp. 85–111, Jan. 2021

    2021

  35. [42]

    Applications of deep reinforcement learning in communications and networking: A survey,

    N. C. Luong, D. T. Hoang, S. Gong, D. Niyato, P. Wang, Y .-C. Liang, and D. I. Kim, “Applications of deep reinforcement learning in communications and networking: A survey,”IEEE Commun. Surv. Tutor., vol. 21, no. 4, pp. 3133–3174, Fourthquarter 2019

    2019

  36. [43]

    Channel estimators for full-duplex communication using orthogonal pilot sequences,

    A. M. M. Chandran, L. Wang, and M. Zawodniok, “Channel estimators for full-duplex communication using orthogonal pilot sequences,”IEEE Access, vol. 8, pp. 117 706–117 713, Jun. 2020

    2020

  37. [44]

    Maxi- mum likelihood localization method with MIMO-OFDM transmission,

    W. A. Mahyiddin, A. L. A. Mazuki, K. Dimyati, and F. Erman, “Maxi- mum likelihood localization method with MIMO-OFDM transmission,” IEEE Access, vol. 9, pp. 150 355–150 365, Nov. 2021

    2021

  38. [45]

    Review and perspective for distance-based clustering of vehicle trajectories,

    P. C. Besse, B. Guillouet, J.-M. Loubes, and F. Royer, “Review and perspective for distance-based clustering of vehicle trajectories,”IEEE Trans. Intell. Transp. Syst., vol. 17, no. 11, pp. 3306–3317, Nov. 2016

    2016

  39. [46]

    Deep reinforcement learning: A brief survey,

    K. Arulkumaran, M. P. Deisenroth, M. Brundage, and A. A. Bharath, “Deep reinforcement learning: A brief survey,”IEEE Signal Process. Mag., vol. 34, no. 6, pp. 26–38, Nov. 2017

    2017

  40. [47]

    Deep reinforcement learning based resource allocation for network slicing with massive mimo,

    D. Yan, B. K. Ng, W. Ke, and C.-T. Lam, “Deep reinforcement learning based resource allocation for network slicing with massive mimo,”IEEE Access, vol. 11, pp. 75 899–75 911, Jul. 2023

    2023

  41. [48]

    Quantifying generalization in reinforcement learning,

    K. Cobbe, O. Klimov, C. Hesse, T. Kim, and J. Schulman, “Quantifying generalization in reinforcement learning,” inInt. Conf. Mach. Learn. PMLR, 2019, pp. 1282–1289

    2019

  42. [49]

    Stochastic geometry modeling and analysis of single- and multi-cluster wireless networks,

    S. M. Azimi-Abarghouyi, B. Makki, M. Haenggi, M. Nasiri-Kenari, and T. Svensson, “Stochastic geometry modeling and analysis of single- and multi-cluster wireless networks,”IEEE Trans. Commun., vol. 66, no. 10, pp. 4981–4996, Oct. 2018

    2018

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.