Pith. sign in

REVIEW 3 major objections 6 minor 19 references

FSIM: Fluid Element Stacked Intelligent Metasurface for Multiuser Downlink Networks

T0 review · 3 major / 6 minor · reviewed 2026-08-01 · deepseek-v4-flash

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

Pith's one-line read A stacked intelligent metasurface whose meta-atoms can slide within each layer outperforms fixed-element SIMs in multiuser downlink sum-rate.

desk verdict FSIM concept is fresh, but the phase-shift subproblem rests on an invalid identity that undermines the simulation results. read the letter →

arxiv 2607.15651 v1 pith:R4HESQUM submitted 2026-07-17 eess.SP

Tsung-Yu Wei , Li-Hsiang Shen , Kai-Ten Feng , Lie-Liang Yang This is my paper · ORCID
classification eess.SP
keywords StackedintelligentmetasurfaceFluidelementMovablemeta-atomsSum-ratemaximizationAlternatingoptimizationSuccessiveconvexapproximationMultiuserMISONear-fieldbeamforming
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 adding positional freedom to a stacked intelligent metasurface (SIM) meaningfully improves multiuser downlink performance. The authors propose FSIM, where the meta-atoms in each layer move within a two-dimensional fluidic region, and jointly optimize element positions, base-station beamforming, and phase-shifts. They show through simulation that this joint position-plus-phase control yields higher sum-rate than conventional fixed-position SIMs, patch-based structures, partially movable designs, and existing flexible intelligent metasurfaces. The practical payoff, if true, is that a compact multi-layer surface with movable elements can better shape cascaded channels toward users and away from interference.

What carries the argument

The load-bearing object is the cascaded near-field channel model S = Φ_L W_L ... Φ_2 W_2 Φ_1, where each W_l is a matrix of free-space spherical-wave coupling coefficients between adjacent layers. The key trick is that moving a meta-atom changes only the geometric distance and direction in W_l, so position becomes an explicit optimization variable with closed-form derivatives. The solution machinery is alternating optimization: beamforming and phase-shift subproblems are relaxed to semidefinite programs solved by successive convex approximation plus rank-one penalty, while the position subproblem uses projected gradient ascent.

What would settle it

Measure or full-wave-simulate the transmission between two FSIM layers as meta-atoms are physically displaced; if the observed inter-layer coupling deviates from the free-space spherical-wave formula in equation (2), the claimed sum-rate advantage of fluid positioning cannot be realized.

Watch

Extended reading notes

Core claim

The paper's central claim is that letting meta-atoms move within each layer of a stacked intelligent metasurface adds a new degree of freedom that phase-only control cannot match. By reformulating the cascaded channel S = Φ_L W_L ... Φ_2 W_2 Φ_1, where each W_l depends on element geometry, the authors show that element positions directly reshape the inter-layer coupling and therefore the end-to-end channel. An alternating optimization algorithm updates beamforming, phase-shifts, and positions, with position updates using the gradient of sum-rate with respect to element coordinates. Simulation results report that full-boundary fluidity achieves the highest rates among all compared fluid archi

Load-bearing premise

The inter-layer coupling model treats each displaced meta-atom as a point-like scatterer whose only effect is a changed propagation distance; if actual movement alters element patterns, mutual coupling, or parasitic responses, the computed sum-rates overstate the real benefit.

Editorial extensions

If this is right

  • If FSIM works as modeled, full-boundary element mobility across all layers is the most valuable configuration, beating partial, patch-limited, single-axis, and discrete-port fluid designs.
  • A compact FSIM with small inter-layer thickness is preferable, because thicker layers weaken inter-layer coupling and reduce the achievable sum-rate.
  • The update order matters: optimizing positions before beamforming and phase-shifts (P-B-T) yields the highest sum-rate in simulations.
  • The proposed alternating optimization algorithm converges and outperforms metaheuristic methods, sub-schemes with disabled variables, and conventional beamforming benchmarks.
  • More layers and more elements increase sum-rate, especially in the near-field region, because they add finer-grained spatial degrees of freedom.

Reading between the lines

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

  • Editorial inference: The result depends on the assumption that moving a meta-atom changes only the geometric distance in W_l while leaving element patterns and mutual coupling unchanged; if real fluid elements alter their electromagnetic response when displaced, the simulated gains could shrink.
  • Editorial inference: The position-gradient update could be extended to dynamic user locations, allowing the FSIM to track users by continuously adjusting element positions rather than only phases.
  • Editorial inference: The same position-optimization idea might apply to other layered reconfigurable surfaces where element placement, not just phase, can be tuned, potentially improving interference suppression in dense networks.
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 / 6 minor

Summary. The paper proposes a fluid-element stacked intelligent metasurface (FSIM) architecture for multiuser downlink MISO networks. Each metasurface layer's meta-atoms can move within a two-dimensional fluidic region, adding positional degrees of freedom to the conventional phase-only SIM. The authors formulate a sum-rate maximization problem jointly over transmit beamforming, phase shifts, and meta-atom positions. They develop an alternating optimization (AO) algorithm: the beamforming and phase-shift subproblems are handled via semidefinite programming with successive convex approximation and penalty-based rank-one relaxation, while the position subproblem is addressed by a projected gradient method. Simulation results claim the proposed FSIM outperforms fixed-position SIMs, partially fluid structures, patch-based designs, and benchmark algorithms including GA, PSO, ZF, and MMSE.

Significance. If the claimed results were correct, the paper would make a useful contribution by showing that moving meta-atoms within SIM layers yields meaningful sum-rate gains beyond phase-only reconfiguration. The system model, with a layered near-field channel and Rician user links, is clearly presented, and the comparisons against external benchmarks (GA/PSO/ZF/MMSE) are appropriate. The paper ships no code or proofs, but the algorithmic structure is standard. However, the central phase-shift reformulation contains a serious algebraic error: the identity used to derive the phase-shift subproblem is wrong for non-diagonal A_l. This means the reported phase updates optimize a different objective than the stated sum-rate, so the simulation results do not, as they stand, support the headline claims. The position projection in (23) also does not demonstrably enforce the physical spacing constraints. These are load-bearing issues that require correction and re-evaluation of the numerical results.

major comments (3)
  1. [III-B, Eq. (12) and the definition of q_{k,k}] The expansion |g_k p_k|^2 = |ψ_l^H q_{k,k}|^2 is invalid when A_l is non-diagonal. From (3), g_k p_k = \bar g_k^H A_l diag(ψ_l) C_l W_1 p_k. Setting a=\bar g_k, b=W_1 p_k, we have a^H A_l diag(ψ_l) C_l b = ψ_l^T[(C_l b) \odot (A_l^T a^*)], not ψ_l^H diag(a^*) A_l C_l b. For L=1, N=2, a=[1,j]^T, W_1 p=[1,1]^T, ψ=[1,j]^T, the true |g p|^2 is 4 while |ψ^H q|^2 is 0. Therefore (12)-(15) and Algorithm 1, step 4, optimize a different objective from the sum-rate in (5). This error must be corrected and the simulations rerun before the performance claims can be assessed.
  2. [III-C, Eq. (23)] The projection Π_XFE in (23) is not a valid projection onto the feasible set RX defined by (1a)-(1e). The formula clamps each coordinate using a single neighboring coordinate, but the minimum-distance constraints (1a)-(1b) involve pairwise spacings across the entire layer. No argument is given that iterates satisfy all spacing constraints simultaneously, and no numerical check is reported. Since the optimized positions are used in every simulation, the physical feasibility of the reported configurations is not established.
  3. [Section IV and Algorithm 1] Because of the phase-shift reformulation error, all simulated sum-rate values attributed to the proposed algorithm are not maximizing the stated objective. Consequently, the comparisons in Figs. 1-4 and the abstract's conclusion that the proposed FSIM 'significantly outperforms' prior designs are unsupported by the present manuscript. Additionally, the penalty-SCA loop has no convergence proof; the paper should report the final penalty values or rank metrics to show that the rank-one relaxation is tight for the reported results.
minor comments (6)
  1. [Eq. (2)] The distance expression contains a typo: it should be sqrt( d_layer^2 + (x^l_{n_h,n_v}-x^{l-1}_{n_h,n_v})^2 + (z^l_{n_h,n_v}-z^{l-1}_{n_h,n_v})^2 ). The minus sign is missing in the z-term.
  2. [III-A, Eq. (11) vicinity] The gradient ∇z_k^T(F^{(t)}) is written with a factor '· ln 2' in the numerator; it should be divided by ln 2, since z_k is a log2 function. This is likely a typographical error but should be corrected.
  3. [III-B definitions] The definitions of A_l and C_l are ambiguous. The text should explicitly state the product indices and that S = A_l diag(ψ_l) C_l, otherwise the decomposition cannot be followed.
  4. [Fig. 1] The observation that ZF performs worse than the random baseline is surprising and should be explained; without justification, it weakens confidence in the benchmark setup.
  5. [Fig. 3] The label 'FSIM as TX (5λ)' is not defined in the text. Clarify what this benchmark represents.
  6. [II, Eq. (1e)] Constraint (1e) forces all meta-atoms on a layer to share the same y-coordinate, which is inconsistent with the preceding statement that they move within a two-dimensional panel. Clarify the geometry.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the FSIM derivation is self-contained and the comparisons are either external or feasible-set-dominance checks.

full rationale

The paper's central claim is a simulation-based sum-rate comparison, not a fitted prediction. The sum-rate objective is stated directly in (5)-(6), and each AO block (beamforming in (11), phase-shift in (15), positions in (23)) optimizes that same objective or a surrogate; no parameter is fitted to the benchmark outputs and then relabeled as a prediction. The comparisons in Section IV include genuinely external benchmarks (ZF, MMSE, GA, PSO) and constrained variants of the same feasible set (fixed-position, partial-fluidity, restricted-region cases); the latter are expected to be dominated by the full-feasible-set formulation, which is a consistency check rather than a circular derivation. The paper's self-citations ([2], [10], [12], [13], [18]) appear in related-work or baseline contexts and are not load-bearing for the FSIM channel model, the algorithm, or the performance conclusion. One reviewer-visible algebraic concern exists in Sec. III-B: the expansion q_{k,k} = diag(¯g_k^H) A_l C_l W_1 p_k and the identity |g_k p_k|^2 = |ψ_l^H q_{k,k}|^2 are not generally valid for non-diagonal A_l, since A_l diag(ψ_l) does not commute. If sustained, this is a correctness defect in the phase subproblem, not a circularity: it means (15) may optimize a different function, but it does not make the output equal to an input by construction. Overall, no load-bearing circular step was found.

Assumptions & free parameters 2 free parameters · 5 assumptions · 1 invented entities

The central performance claim rests on a self-consistent but unverified channel model, on algorithm hyperparameters that are not reported, and on a proposed FSIM architecture with no independent evidence. The simulation comparisons are against benchmarks generated from the same model, so the main uncharged input is the near-field coupling model rather than fitted data.

free parameters (2)
  • Algorithm hyperparameters (mu1, mu2, alpha1, alpha2, eta, epsilon, Tmax) = not reported
    Penalty and gradient steps control convergence of the SCA/rank-one relaxation and projected gradient; no values or sensitivity analysis are given.
  • Simulation scenario constants (user radius, Rician factor kappa=10, path-loss exponent 2.5, geometry) = specified in Section IV
    Performance ordering may depend on these choices; no sweeps are shown except transmit power, number of layers, and number of elements.
assumptions (5)
  • domain assumption Near-field inter-layer propagation follows the spherical-wave model in eq. (2) with scalar phase per meta-atom and no explicit mutual coupling beyond W_l.
    Invoked in Section II; the entire cascaded channel and all position gradients depend on this formula.
  • domain assumption The FSIM cascade is modeled as point-wise phase shifts multiplied layer-by-layer in eq. (3).
    This ignores coupling between phase tuning and element movement except through the geometric W_l matrices.
  • domain assumption User links follow Rician fading with sinc spatial correlation in eq. (4).
    Simulation-only assumption; no measurements support this model for the FSIM last-layer-to-user channel.
  • domain assumption The BS-FSIM channel W1 follows the same near-field coupling structure, though no explicit expression is provided.
    W1 is introduced in Section II but never defined; reproduction requires choosing a model.
  • ad hoc to paper The penalty-SCA and projected-gradient updates converge to useful local solutions.
    Algorithm 1 asserts convergence via Fig. 1; no convergence or optimality proof is given, and the projection in (23) is not demonstrated to satisfy all spacing constraints.
invented entities (1)
  • Fluid-element stacked intelligent metasurface (FSIM)
    purpose: A layered metasurface whose meta-atoms can move within each layer to add spatial diversity beyond phase control.
    Proposed system concept evaluated only via simulation; no prototype, measurement, or independent hardware evidence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FSIM: Fluid Element Stacked Intelligent Metasurface for Multiuser Downlink Networks." pith.science (2026). https://pith.science/paper/R4HESQUM

@misc{pith2026260715651,
  author       = {Pith},
  title        = {Pith review of: FSIM: Fluid Element Stacked Intelligent Metasurface for Multiuser Downlink Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/R4HESQUM}},
  note         = {Machine review of arXiv:2607.15651}
}
read the original abstract

A fluid element (FE)-aided stacked intelligent metasurface (FSIM) for multiple-input single-output (MISO) communication system is investigated, where a multi-antenna base station (BS) serves multiple single-antenna users through FSIM. Unlike conventional SIM with fixed meta-atom deployment, the architecture allows the meta-atoms in each layer to move within a predefined fluidic region to further increase the spatial diversity. By jointly optimizing the two-dimensional meta-atom positions, BS transmit beamforming, and FSIM phase-shifts, the cascaded BS-FSIM-user channels can be flexibly reconfigured to enhance the desired signals and suppress multiuser interference. The proposed sum-rate maximization problem is highly non-convex and nonlinear due to the coupled solutions of element position, beamforming, and phase-shift. To address this challenge, an alternating optimization (AO) algorithm is developed to iteratively update these variables. The beamforming and FSIM phase-shift subproblems are transformed into semi-definite programming problems and solved by using successive convex approximation (SCA), first-order Taylor approximation, and penalty-based rank-one relaxation, whilst the FE position subproblem is handled through a projected gradient-based update. Simulation results reveal that a compact FSIM with a small inter-layer thickness is preferable, as increasing the thickness weakens inter-layer coupling and degrades the achievable sum rate. Results also demonstrate that the proposed FSIM significantly outperforms conventional SIMs with fixed positions, patch-based structures, partial fluidity, restricted fluid regions, and existing flexible intelligent metasurfaces. Furthermore, the proposed AO-based algorithm achieves superior rate performance compared to sub-schemes, metaheuristic methods, and conventional beamforming benchmarks.

Figures

Figures reproduced from arXiv: 2607.15651 by the authors.

Figure 1
Figure 1. Convergence. 10 20 30 40 Transmit Power (dBm) 10 15 20 25 Sum-Rate (bps/Hz) M = 4, K = 4 M = 10, K = 4 M = 16, K = 4 M = 4, K = 2 M = 10, K = 2 M = 16, K = 2 M = 4, K = 6 M = 10, K = 6 M = 16, K = 6 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references

  1. [1]

    Smart radio environments empowered by reconfig- urable intelligent surfaces: How it works, state of researc h, and the road ahead,

    M. D. Renzo et al., “Smart radio environments empowered by reconfig- urable intelligent surfaces: How it works, state of researc h, and the road ahead,” IEEE J. Sel. Areas Commun. , vol. 38, no. 11, pp. 2450–2525, 2020

    2020

  2. [2]

    Joint active and passive beamforming for energy- efficient STARS with quantization and element selection in I SAC systems,

    L.-H. Shen et al. , “Joint active and passive beamforming for energy- efficient STARS with quantization and element selection in I SAC systems,” IEEE Trans. Commun. , vol. 74, pp. 1003–1018, 2026

    2026

  3. [3]

    Stacked intelligent metasurfaces for efficient holograp hic MIMO communications in 6G,

    J. An et al. , “Stacked intelligent metasurfaces for efficient holograp hic MIMO communications in 6G,” IEEE J. Sel. Areas Commun. , vol. 41, no. 8, pp. 2380–2396, 2023

    2023

  4. [4]

    Stacked intelligent metasurfaces for multiuser do wnlink beam- forming in the wave domain,

    ——, “Stacked intelligent metasurfaces for multiuser do wnlink beam- forming in the wave domain,” IEEE Trans. Wireless Commun. , vol. 24, no. 7, pp. 5525–5538, 2025

    2025

  5. [5]

    Performance limits of fluid antenna systems,

    K.-K. Wong et al., “Performance limits of fluid antenna systems,” IEEE Commun. Lett. , vol. 24, no. 11, pp. 2469–2472, 2020

    2020

  6. [6]

    Fluid antenna systems,

    ——, “Fluid antenna systems,” IEEE Trans. Wireless Commun., vol. 20, no. 3, pp. 1950–1962, 2021

    1950

  7. [7]

    Fluid antenna multiple access,

    ——, “Fluid antenna multiple access,” IEEE Trans. Wireless Commun. , vol. 21, no. 7, pp. 4801–4815, 2022

    2022

  8. [8]

    On performance of RIS-aided fluid antenna systems,

    F. R. Ghadi et al., “On performance of RIS-aided fluid antenna systems,” IEEE Wireless Commun. Lett. , vol. 13, no. 8, pp. 2175–2179, 2024

    2024

Show all 19 references
  1. [9]

    Movable-element RIS-aided wireless communications: An element-wise position optimization approach,

    J. Zhao et al. , “Movable-element RIS-aided wireless communications: An element-wise position optimization approach,” IEEE Commun. Lett. , vol. 30, pp. 967–971, 2026

    2026

  2. [10]

    6D movable metasurface (6DMM) in downlink NOMA transmissions,

    L.-H. Shen, “6D movable metasurface (6DMM) in downlink NOMA transmissions,” IEEE Commun. Lett. , vol. 30, pp. 517–521, 2026

    2026

  3. [11]

    Movable-element RIS: Joint element positioning and beamforming optimization,

    D. Zhou et al. , “Movable-element RIS: Joint element positioning and beamforming optimization,” IEEE Wireless Commun. Lett. , vol. 15, pp. 915–919, 2025

    2025

  4. [12]

    RIS-aided fluid antenna array-mounted AA V net- works,

    L.-H. Shen et al. , “RIS-aided fluid antenna array-mounted AA V net- works,” IEEE Wireless Commun. Lett. , vol. 14, no. 4, pp. 1049–1053, 2025

    2025

  5. [13]

    Liquid intelligent metasurface for fluid a ntennas-assisted networks,

    L.-H. Shen, “Liquid intelligent metasurface for fluid a ntennas-assisted networks,” IEEE Wireless Commun. Lett. , vol. 14, no. 12, pp. 4192– 4196, 2025

    2025

  6. [14]

    RIS-aided wireless communication with movable elements: Geometry impact on performance,

    Y . Zhang et al. , “RIS-aided wireless communication with movable elements: Geometry impact on performance,” in Proc. IEEE Int. Conf. Electr . Comput. Energy Technol. (ICECET) , 2025, pp. 1–6

    2025

  7. [15]

    Movable antenna-enabled RIS-aided integrated sensing and communication,

    H. Wu et al. , “Movable antenna-enabled RIS-aided integrated sensing and communication,” IEEE Trans. Cogn. Commun. Netw. , vol. 11, no. 5, pp. 2879–2892, 2025

    2025

  8. [16]

    Fluid reconfigurable intelligent surfaces: Joint on-off selection and beamforming with discrete phase shifts,

    H. Xiao et al. , “Fluid reconfigurable intelligent surfaces: Joint on-off selection and beamforming with discrete phase shifts,” IEEE Wireless Commun. Lett. , vol. 14, no. 10, pp. 3124–3128, 2025

    2025

  9. [17]

    Rayleigh fading modeling and channel hardening for reconfigurable intelligent surfaces,

    E. Bj¨ ornson et al. , “Rayleigh fading modeling and channel hardening for reconfigurable intelligent surfaces,” IEEE Wireless Commun. Lett. , vol. 10, no. 4, pp. 830–834, 2021

    2021

  10. [18]

    Mobility-aware subband and beam resource allocation schemes for millimeter wave wireless networks,

    L.-H. Shen et al., “Mobility-aware subband and beam resource allocation schemes for millimeter wave wireless networks,” IEEE Trans. V eh. Technol., vol. 69, no. 10, pp. 11 893–11 908, Oct. 2020. 6

    2020

  11. [19]

    Multiuser communications with movable-antenna base station: Joint antenna positioning, receive combining, an d power con- trol,

    Z. Xiao et al. , “Multiuser communications with movable-antenna base station: Joint antenna positioning, receive combining, an d power con- trol,” IEEE Trans. Wireless Commun., vol. 23, no. 12, pp. 19 744–19 759, 2024

    2024

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.