Pith. sign in

REVIEW 3 major objections 4 minor 22 references

Dual-Polarization Stacked Intelligent Metasurfaces for Holographic MIMO

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper claims that independently controlling two orthogonal polarizations on each stacked metasurface layer makes the transceiver behave like two isolated single-polarization SIMs, doubling the interference-free data streams it can…

desk verdict A plausible stacked dual-polarized metasurface idea, undermined as printed by a channel model typo that erases the polarization-mixing parameter and by a partly circular upper bound; still worth refereeing after corrections. read the letter →

arxiv 2505.20805 v1 pith:W4VYQXWZ submitted 2025-05-27 eess.SP

Yida Zhang , Qiuyan Liu , Hongtao Luo , Yuqi Xia , Qiang Wang This is my paper · ORCID
classification eess.SP
keywords dual-polarizedstackedintelligentmetasurfacesholographicMIMOreconfigurablesurfaceswave-domainsignalprocessingspectralefficiencyenergypolarizationcross-interferenceLGD-WFalgorithm
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 claims that a holographic MIMO transceiver can double its wave-domain signal-processing capacity without adding layers or unit cells by letting each metasurface layer control two orthogonal polarizations independently. The proposed dual-polarized stacked intelligent metasurface (DPSIM) is mathematically equivalent to two isolated single-polarization SIMs, so each polarization path can carry its own data streams while polarization cross-interference and inter-stream interference are suppressed. A layer-by-layer gradient-descent with water-filling algorithm (LGD-WF) tunes the phase shifts so the end-to-end channel approaches the target diagonal structure, allowing spatial multiplexing without digital precoding. Simulations show that, for the same number of layers and unit size, DPSIM supports more simultaneous interference-free streams than a conventional SIM and brings spectral and energy efficiency close to the theoretical upper bound even when polarization conversion is present.

What carries the argument

The central object is the DPSIM transfer-matrix pair in Eqs. (5)-(6): interleaved per-layer diagonal phase matrices $\Phi_p^l$ and $\Psi_p^k$ for the two polarizations $p\in\{0,1\}$ with Rayleigh-Sommerfeld free-space propagation matrices $V^l$ and $U^k$. Because both kinds of matrices are block-diagonal across polarizations, the whole stack factorizes into two isolated SIMs. The LGD-WF algorithm differentiates the fitting objective $\Gamma=\|\alpha RGT-\Lambda_{1:2S,1:2S}\|_F^2$ with respect to each per-polarization phase, normalizes the gradients to avoid explosion or vanishing, updates the phases layer by layer with a decaying learning rate, and finishes with water-filling power allocation across the $2S$ streams.

What would settle it

Run a full-wave simulation of two adjacent dual-polarized metasurface layers and compute the complete $2M \times 2M$ transmission matrix; if the off-diagonal polarization blocks have norm comparable to the diagonal blocks, for example at oblique incidence or with imperfect unit-cell isolation, then the block-diagonal model of Eqs. (5)-(6) fails and the claimed equivalence to two isolated SIMs, together with the per-polarization gradient updates, no longer holds.

Watch

Extended reading notes

Core claim

The central claim is that a stacked intelligent metasurface made of dual-polarized units, with independent phase control per polarization, splits into two parallel, isolated single-polarization SIMs. Because both the per-layer phase matrices and the inter-layer propagation matrices are block-diagonal in the polarization index, the overall transmitter and receiver transfer matrices factor into two identical single-polarization paths; the end-to-end channel can then be driven toward the truncated singular-value matrix of the physical channel by per-polarization gradient updates. At fixed layer count and unit-cell count this yields up to twice as many interference-free data streams as a single-polarization SIM, and the spectral and energy efficiency follow the theoretical upper bound closely across the simulated polarization-imperfection levels.

Load-bearing premise

The argument assumes that a wave keeps its polarization state while travelling between metasurface layers, so the inter-layer propagation matrices are block-diagonal and the two polarizations never mix inside the stack; if polarization conversion between layers is significant, the two paths couple and the system is no longer equivalent to two isolated SIMs.

Editorial extensions

If this is right

  • For a fixed physical footprint, DPSIM supports more simultaneous data streams with negligible inter-stream interference than a single-polarization SIM, because each of the two polarization paths provides an independent parallel channel.
  • The LGD-WF algorithm converges quickly, within about 20 iterations in the simulations, and produces end-to-end channel matrices close to the target diagonal form for both SIM and DPSIM.
  • Increasing the number of metasurface layers improves channel fitting and interference suppression, but DPSIM reaches the same or better fitting at the same layer count, leaving integration space for other functions.
  • Spectral and energy efficiency of DPSIM-assisted HMIMO approach the theoretical upper bound for polarization conversion power ratios of 0.2 and 0.4, indicating robustness to polarization cross-interference.
  • DPSIM-assisted HMIMO achieves higher energy efficiency than conventional massive MIMO because it avoids digital precoding and uses low-resolution data converters.

Reading between the lines

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

  • Because the stack factorizes by polarization, one could run the LGD-WF updates independently on each polarization block, roughly halving the optimization dimension; the paper does not explicitly use this computational saving.
  • If measurable polarization conversion occurs between layers, the block-diagonal model would need to be replaced by full $2M \times 2M$ inter-layer blocks; the same gradient framework could still be applied to all four polarization blocks, trading the clean doubling for robustness.
  • The DPSIM substitution is a drop-in structural upgrade, so the same idea may carry over to other SIM-based systems such as multiuser beamforming, integrated sensing and communication, and semantic communication, potentially doubling their wave-domain degrees of freedom.
  • A direct testable consequence of the equivalence claim is that a DPSIM with $M$ dual-polarized units per layer should match the channel-fitting quality of a single-polarization SIM with $2M$ units per layer under the same optimization budget.
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

3 major / 4 minor

Summary. The paper proposes a dual-polarized stacked intelligent metasurface (DPSIM) architecture for holographic MIMO, in which each metasurface layer independently controls the phase of two orthogonal polarizations. The authors model the end-to-end channel, formulate an optimization problem to shape the effective channel toward the singular-value matrix of the physical channel, and propose a layer-by-layer gradient descent with water-filling (LGD-WF) algorithm. Simulations claim that, for the same number of layers and unit size, DPSIM supports more interference-free streams and achieves higher spectral and energy efficiency than single-polarization SIM and conventional massive MIMO, approaching a theoretical upper bound.

Significance. The core idea—using independent phase control of two polarizations to double the wave-domain processing capability of a stacked metasurface within the same physical footprint—is timely and plausible, and the paper provides a concrete architecture, an iterative algorithm, and a public simulation code link. If the claims were fully supported, this would be a useful extension of the SIM literature. However, the quantitative evidence is currently undermined by a degenerate printed channel model, a circular performance bound, and an energy-efficiency metric that omits all hardware power consumption. The qualitative advantage of DPSIM may survive a corrected analysis, but the present manuscript does not establish it.

major comments (3)
  1. [§II, Eq. (8)] As printed, the channel model in Eq. (8) makes the four polarization blocks of G identical and removes all dependence on the polarization conversion ratio ε. With A = R_RX^(1/2) and B = R_TX^(1/2), the pre- and post-multipliers are [A A; A A] and [B B; B B], so each block of G becomes A(G00+G10+G01+G11)B. Since G00,G11 have variance (1−ε)PL and G10,G01 have variance εPL, the sum has variance 2PL, independent of ε. Consequently, the ε-dependent SE/EE curves in Fig. 6 and the claim of robustness to polarization imperfections cannot follow from the stated model. The likely intended form is diag(R_RX^(1/2), R_RX^(1/2)) · [G00 G01; G10 G11] · diag(R_TX^(1/2), R_TX^(1/2)); please correct Eq. (8) and rerun the simulations, or explain explicitly why the printed form is not what was implemented.
  2. [§III, P1 and §IV, Eq. (27)] The design target Λ in P1 is the singular-value matrix of the physical channel G, and the theoretical upper bound η_SE^ub in Eq. (27) is computed from exactly the same Λ, used also for water-filling power allocation. Therefore the reported 'approach to the theoretical upper bound' measures only how well LGD-WF fits the algorithm's own target, not how close DPSIM comes to any fundamental or architectural limit. This circularity should be acknowledged, and the upper-bound claim should be reframed as, at best, a check of the optimization algorithm's convergence to the chosen target.
  3. [§IV, Eq. (28)] The energy efficiency definitions in Eq. (28) divide spectral efficiency only by the transmit power P_t, omitting all hardware power consumption: RF chains, DACs/ADCs, metasurface control, and baseband processing. The claim that DPSIM-assisted HMIMO has 'significantly higher EE' than 256×32 and 512×32 massive MIMO systems is therefore not a systems-level energy comparison. Please include a total power model, or relabel the metric as transmit-power-normalized SE and temper the EE conclusions accordingly.
minor comments (4)
  1. [§II, Eqs. (5)–(6)] The stated dimensions are inconsistent: the products in (5) and (6) are built from 2M×2M and 2N×2N block-diagonal matrices, yet T is declared as C^(2M×2S) and R as C^(2S×2N). Please specify how the S active streams are selected from the M or N units (e.g., a truncation or an explicit input/output mapping).
  2. [§III, Step 3] There is a typo in the sentence preceding Eqs. (20)–(21): 'Our phase update strategies for TX-DPSIM and TX-DPSIM' should presumably read 'TX-DPSIM and RX-DPSIM'.
  3. [§IV, Eq. (24)] The water-filling expression diag(Λ_{1:2S,1:2S}Λ_{1:2S,1:2S}) is confusing; it should be written as the squared singular values, e.g., [Λ]^2_s, to match the SINR terms in Eqs. (26)–(27).
  4. [§I, footnote and code link] The GitHub URL in the footnote contains spaces: 'Dual Polarization Stacked Intelligent Metasurfaces for Holographic MIMO.git'. Please provide a URL-encoded link so that the code is actually accessible.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: DPSIM's stream-count and SE/EE gains are structural consequences of the stated two-polarization model, and the Lambda-based upper bound is an independent benchmark the algorithm must actively approach.

full rationale

The paper's derivations are self-contained; no load-bearing step reduces to its own input by construction. DPSIM is explicitly modeled as two isolated single-polarization SIMs (Section II, Eqs. (5)-(6)), so the larger number of interference-free streams follows structurally from having two independently controllable polarization paths per layer, not from reusing the paper's conclusion. The LGD-WF algorithm (P1, Eq. (13)) minimizes the distance from the end-to-end channel to the truncated SVD target Lambda of G; the NMSE, SE, and EE metrics then measure how well that optimization succeeds. The theoretical upper bound in Eq. (27) is the standard water-filling capacity over G's singular values and is an independent information-theoretic benchmark; the algorithm must actually diagonalize the channel to approach it, so the comparison is not circular. Citations to [4], [12], [16] are external prior works, not self-citations by the present authors, and the block-diagonal form of T and R is a stated propagation assumption. A separate correctness issue, not a circularity, is that Eq. (8)'s block-repeated correlation matrices, combined with Eqs. (9)-(10), make the distribution of G independent of the polarization-conversion ratio epsilon; thus the distinct EE curves for epsilon=0.2 and epsilon=0.4 in Fig. 6 do not follow from the printed model. This is an internal channel-model inconsistency and is outside the circularity score.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim relies on: (i) the block-diagonal polarization model that makes DPSIM equivalent to two SIMs, (ii) a channel model with a hand-set polarization leakage parameter epsilon, and (iii) an unproven iterative optimization. No genuinely new physical entities are introduced.

free parameters (3)
  • polarization conversion power ratio epsilon = 0.2 (default), 0.4 (varied)
    Chosen by hand in Table I and Fig. 6 to model polarization imperfection. The main claim of PCI suppression depends on this parameter being small; the paper does not sweep it over the full range.
  • LGD-WF hyperparameters eta_0, beta, E_max, ς = 0.1, 0.5, 20, 100
    Hand-selected algorithm parameters. They affect convergence speed and final NMSE, and no sensitivity analysis is provided.
  • compensation scaling factor alpha = optimized by least squares (Eq. 22)
    Fit to the channel and current phase configuration to minimize the objective in P1. It enters the reported SE comparison.
assumptions (4)
  • domain assumption Polarization state of EM waves is unchanged between DPSIM layers (Section II, before Eq. (3))
    This justifies the block-diagonal structure of T and R in Eqs. (5)-(6). If violated, the two polarizations couple inside the stack and the DPSIM cannot be treated as two isolated SIMs.
  • standard math Rayleigh-Sommerfeld diffraction formula governs inter-layer transmission (Eqs. (3)-(4))
    Standard propagation model cited from [14]; it is used to build the fixed V_l and U_k matrices.
  • domain assumption The channel correlation model in Eq. (8) with repeated R^{1/2} blocks is the correct dual-polarized spatial correlation model
    The paper cites [12], [16] but Eq. (8) as printed makes all four polarization blocks identical. This is likely a typesetting error, but as written it contradicts the intended separate 00/11/10/01 matrices.
  • ad hoc to paper The gradient descent in LGD-WF converges to a good local optimum (Algorithm 1, no convergence proof)
    The algorithm uses normalized gradient steps with decaying learning rate and random restarts, but no convergence guarantee or complexity analysis is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Dual-Polarization Stacked Intelligent Metasurfaces for Holographic MIMO." pith.science (2026). https://pith.science/paper/W4VYQXWZ

@misc{pith2026250520805,
  author       = {Pith},
  title        = {Pith review of: Dual-Polarization Stacked Intelligent Metasurfaces for Holographic MIMO},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/W4VYQXWZ}},
  note         = {Machine review of arXiv:2505.20805}
}
read the original abstract

To address the limited wave domain signal processing capabilities of traditional single-polarized stacked intelligent metasurfaces (SIMs) in holographic multiple-input multiple-output (HMIMO) systems, which stems from limited integration space, this paper proposes a dual-polarized SIM (DPSIM) architecture. By stacking dual-polarized reconfigurable intelligent surfaces (DPRIS), DPSIM can independently process signals of two orthogonal polarizations in the wave domain, thereby effectively suppressing polarization cross-interference (PCI) and inter-stream interference (ISI). We introduce a layer-by-layer gradient descent with water-filling (LGD-WF) algorithm to enhance end-to-end performance. Simulation results show that, under the same number of metasurface layers and unit size, the DPSIM-aided HMIMO system can support more simultaneous data streams for ISI-free parallel transmission compared to traditional SIM-aided systems. Furthermore, under different polarization imperfection conditions, both the spectral efficiency (SE) and energy efficiency (EE) of the DPSIM-aided HMIMO system are significantly improved, approaching the theoretical upper bound.

Figures

Figures reproduced from arXiv: 2505.20805 by the authors.

Figure 1
Figure 1. Schematic diagram of the DPSIM-assisted HMIMO system. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The schematic diagram of the DPSIM-assisted HMIMO system in [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. NMSE convergence diagram with algorithm iteration. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: End-to-end spatial channel matrix. The path loss between the transmitter and the receiver is modeled by [22] PL(d) = PL(d0) + 10b log10  d d0  + Xδ, d ≥ d0, (29) where PL(d0) = 20 log10 4πd0 λ  dB is the free space path loss at the reference distance d0, b represent…
Figure 5
Figure 5. Figure 5: The HMIMO system’s SE varies with the number of data streams. [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: The HMIMO system’s EE varies with the transmission power change. [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 19 canonical work pages

  1. [1]

    SVD-assisted multiuser transmitter and multiuser detector design for MIMO systems,

    W. Liu, L. L. Yang, and L. Hanzo, “SVD-assisted multiuser transmitter and multiuser detector design for MIMO systems,”IEEE Trans. V eh. Technol., vol. 58, no. 2, pp. 1016–1021, June 2009

    work page 2009

  2. [2]

    An introduction to millimeter-wave mobile broadband systems,

    Z. Pi, “An introduction to millimeter-wave mobile broadband systems,” IEEE Commun. Mag., vol. 49, no. 6, pp. 101–107, June 2011

    work page 2011

  3. [3]

    Holographic MIMO communications: Theoretical foundations, enabling technologies, and future directions,

    T. Gong, P. Gavriilidis, R. Ji, C. Huang, G. C. Alexandropoulos, L. Wei, Z. Zhang, M. Debbah, H. V . Poor, and C. Yuen, “Holographic MIMO communications: Theoretical foundations, enabling technologies, and future directions,”IEEE Commun. Surv. Tuts., vol. 26, no. 1, pp. 196– 257, Aug. 2024

    work page 2024

  4. [4]

    Stacked intelligent metasurfaces for efficient holographic MIMO communications in 6G,

    J. An, C. Xu, D. W. K. Ng, G. C. Alexandropoulos, C. Huang, C. Yuen, and L. Hanzo, “Stacked intelligent metasurfaces for efficient holographic MIMO communications in 6G,”IEEE J. Sel. Areas Commun., vol. 41, no. 8, pp. 2380–2396, June 2023

    work page 2023

  5. [5]

    Stacked intelligent metasurfaces-enhanced MIMO OFDM wideband communication systems,

    Z. Li, J. An, and C. Yuen, “Stacked intelligent metasurfaces-enhanced MIMO OFDM wideband communication systems,” Mar. 2025. [Online]. Available: https://arxiv.org/abs/2503.00368

    arXiv 2025

  6. [6]

    Stacked intelligent metasurfaces for multiuser downlink beamforming in the wave domain,

    J. An, M. D. Renzo, M. Debbah, H. Vincent Poor, and C. Yuen, “Stacked intelligent metasurfaces for multiuser downlink beamforming in the wave domain,”IEEE Trans. Wireless Commun., pp. 1–1, Mar. 2025

    work page 2025

  7. [7]

    Transmit beamforming design for ISAC with stacked intelligent metasurfaces,

    S. Li, F. Zhang, T. Mao, R. Na, and Z. Wang, “Transmit beamforming design for ISAC with stacked intelligent metasurfaces,”IEEE Trans. V eh. Technol., vol. 74, no. 4, pp. 6767–6772, July 2025

    work page 2025

  8. [8]

    Stacked intelligent metasurfaces for task-oriented semantic communications,

    G. Huang, J. An, Z. Yang, L. Gan, M. Bennis, and M. Debbah, “Stacked intelligent metasurfaces for task-oriented semantic communications,” July 2024. [Online]. Available: https://arxiv.org/abs/2407.15053

    arXiv 2024

Show all 22 references
  1. [9]

    Linear and nonlinear polarization syntheses and their programmable controls based on anisotropic time-domain digital coding metasurface,

    J. C. Ke, J. Y . Dai, M. Z. Chen, L. Wang, C. Zhang, W. Tang, J. Yang, W. Liu, X. Li, Y . Lu, Q. Cheng, S. Jin, and T. J. Cui, “Linear and nonlinear polarization syntheses and their programmable controls based on anisotropic time-domain digital coding metasurface,”Small Struct...

    2021

  2. [10]

    Anisotropic coding metamaterials and their powerful manipulation of differently polarized terahertz waves,

    S. Liu, T. J. Cui, Q. Xu, D. Bao, L. Du, X. Wan, W. X. Tang, C. Ouyang, X. Y . Zhou, H. Yuan, H. F. Ma, W. X. Jiang, J. Han, W. Zhang, and Q. Cheng, “Anisotropic coding metamaterials and their powerful manipulation of differently polarized terahertz waves,”Light Sci. Appl., vo...

    2016

  3. [11]

    Polarization-controlled dual-programmable metasurfaces,

    X. G. Zhang, Q. Yu, W. X. Jiang, Y . L. Sun, L. Bai, Q. Wang, C.-W. Qiu, and T. J. Cui, “Polarization-controlled dual-programmable metasurfaces,”Adv. Sci., vol. 7, no. 11, pp. 1–10, Apr. 2020

    2020

  4. [12]

    Dual-polarized RIS-assisted mobile communications,

    Y . Han, X. Li, W. Tang, S. Jin, Q. Cheng, and T. J. Cui, “Dual-polarized RIS-assisted mobile communications,”IEEE Trans. Wireless Commun., vol. 21, no. 1, pp. 591–606, July 2022

    2022

  5. [13]

    Dual-polarized reconfigurable intelligent surface-based antenna for holographic mimo communications,

    S. Zeng, H. Zhang, B. Di, Z. Han, and H. Vincent Poor, “Dual-polarized reconfigurable intelligent surface-based antenna for holographic mimo communications,”IEEE Trans. Wireless Commun., vol. 23, no. 11, pp. 17 339–17 353, Sept. 2024

    2024

  6. [14]

    All-optical machine learning using diffractive deep neural networks,

    X. Lin, Y . Rivenson, N. T. Yardimci, M. Veli, Y . Luo, M. Jarrahi, and A. Ozcan, “All-optical machine learning using diffractive deep neural networks,”Science, vol. 361, no. 6406, pp. 1004–1008, Sept. 2018

    2018

  7. [15]

    Analysis and modeling on co- and cross-polarized urban radio propaga- tion for dual-polarized mimo wireless systems,

    V . Degli-Esposti, V .-M. Kolmonen, E. M. Vitucci, and P. Vainikainen, “Analysis and modeling on co- and cross-polarized urban radio propaga- tion for dual-polarized mimo wireless systems,”IEEE Trans. Antennas Propag., vol. 59, no. 11, pp. 4247–4256, Aug. 2011

    2011

  8. [16]

    Holographic beam- forming for ultra massive MIMO with limited radiation amplitudes: How many quantized bits do we need?

    X. Hu, R. Deng, B. Di, H. Zhang, and L. Song, “Holographic beam- forming for ultra massive MIMO with limited radiation amplitudes: How many quantized bits do we need?”IEEE Commun. Lett., vol. 26, no. 6, pp. 1403–1407, Feb. 2022

    2022

  9. [17]

    Spatially-stationary model for holographic MIMO small-scale fading,

    A. Pizzo, T. L. Marzetta, and L. Sanguinetti, “Spatially-stationary model for holographic MIMO small-scale fading,”IEEE J. Sel. Areas Commun., vol. 38, no. 9, pp. 1964–1979, June 2020

    1964

  10. [18]

    Low-complexity channel estima- tion and passive beamforming for RIS-assisted MIMO systems relying on discrete phase shifts,

    J. An, C. Xu, L. Gan, and L. Hanzo, “Low-complexity channel estima- tion and passive beamforming for RIS-assisted MIMO systems relying on discrete phase shifts,”IEEE Trans. Commun., vol. 70, no. 2, pp. 1245–1260, Feb. 2022

    2022

  11. [19]

    A mathematical theory of communication,

    Shannon and E. C., “A mathematical theory of communication,”Bell Syst. Tech. J., vol. 27, no. 4, pp. 623–656, Oct. 1948

    1948

  12. [20]

    Gradient amplification: An efficient way to train deep neural networks,

    S. Basodi, C. Ji, H. Zhang, and Y . Pan, “Gradient amplification: An efficient way to train deep neural networks,”Big Data Min. Anal., vol. 3, no. 3, pp. 196–207, Sept. 2020

    2020

  13. [21]

    Deep learning,

    Y . LeCun, Y . Bengio, and G. Hinton, “Deep learning,”Nature, vol. 521, pp. 436–444, May 2015

    2015

  14. [22]

    Wide- band millimeter-wave propagation measurements and channel models for future wireless communication system design,

    T. S. Rappaport, G. R. MacCartney, M. K. Samimi, and S. Sun, “Wide- band millimeter-wave propagation measurements and channel models for future wireless communication system design,”IEEE Trans. Commun., vol. 63, no. 9, pp. 3029–3056, Sept. 2015

    2015

Pith tools

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