Pith. sign in

REVIEW 3 major objections 5 minor 15 references

Active RIS Enabled NLoS LEO Satellite Communications: A Three-timescale Optimization Framework

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 An active RIS whose orientation, reflection coefficients, and transmit beams are updated on three different clocks can keep a blocked LEO satellite link alive within an energy budget while cutting beam-switching costs.

desk verdict Solid idea, but the energy constraint that drives the claimed savings is algebraically wrong, so the main result isn't supported yet. read the letter →

arxiv 2506.20424 v1 pith:AUNDF553 submitted 2025-06-25 eess.SP

Ziwei Liu , Junyan He , Shanshan Zhao , Meng Hua , Bin Lyu , Xinjie Zhao , Gengxin Zhang This is my paper
classification eess.SP
keywords activereconfigurableintelligentsurfaceLEOsatellitecommunicationsnon-line-of-sightlinksthree-timescaleoptimizationbeamformingswitchingenergyfractionalprogrammingalternatingsuccessiveconvexapproximation
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

An active RIS—a reflective array that amplifies as well as phase-shifts incident signals—can restore a reliable downlink from a low-Earth-orbit satellite to a user in a canyon or dense urban area where the direct path is blocked. This paper tries to show that the best way to run such a system is to optimize the RIS at three different timescales: aim its direction once for the whole pass, update its reflection coefficients at each holding interval, and let the satellite transmit beamforming change every second. Under a total energy budget, lengthening the holding interval saves the switching-and-control energy of the active RIS but sacrifices adaptivity to the fast-changing satellite channel, so the paper's outer-layer search picks the rate-maximizing interval. The authors' two-layer algorithm, built from fractional programming, alternating optimization, successive convex approximation, and a penalty method, is shown by simulation to beat both partial-optimization designs and a semidefinite-relaxation baseline, and to outperform passive RIS with fewer elements. If these results hold, active RIS becomes a plausible energy-constrained enabler for NLoS satellite links rather than a device requiring constant reconfiguration.

What carries the argument

The machinery is the three-timescale schedule itself: the RIS direction vector $\mathbf{r}$ is fixed for the whole communication period, the active-RIS coefficient matrices $\{\Theta_b\}$ are updated once per holding interval, and the satellite transmit beams $\{\mathbf{w}_{b,k}\}$ are updated each $\delta=1$ s slot. The holding interval length $K$ is the control knob, because it appears in the energy constraint as the multiplier of the switching-and-control term $BMNP_C$. Inside each candidate $K$, the inner loop solves three subproblems: transmit beams by Rayleigh quotient, RIS coefficients by fractional programming followed by a penalized rank-one semidefinite program, and RIS orientation by the same penalty-plus-successive-convex-approximation treatment. The outer loop then steps $K$ downward until the energy budget binds and chooses the feasible $K$ with the highest average rate.

What would settle it

Build or measure an active-RIS prototype and log the actual power drawn while following the holding-interval schedule returned by Algorithm 1; if total energy does not match the model expression used in constraint C4, the predicted optimum for $K$ and the claimed energy savings are not grounded.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the RIS direction $\mathbf{r}$, the active-RIS coefficient matrices $\{\Theta_b\}$, and the satellite transmit beams $\{\mathbf{w}_{b,k}\}$ can be co-designed on three separate clocks and still yield a feasible near-optimal configuration, whereas treating them on a single clock or fixing one of them degrades the average rate. The received signal carries the direction via the radiation-pattern factors $\mathbf{r}^T\mathbf{l}_{sr}$ and $\mathbf{r}^T\mathbf{l}_{ru}$, and the energy constraint couples all three clocks through the per-holding-interval switching term $BMNP_C$. The inner-loop algorithm maximizes the average achievable rate $\frac{1}{BK}\sum_{b,k}\log_2(1+\gamma_{b,k})$ for a fixed holding-interval length $K$, and the outer loop decreases $K$ until the energy budget binds, selecting the largest feasible $K$. Simulation comparisons against SDR/GR and partially optimized schemes are the evidence the authors offer for the claim.

Load-bearing premise

The riskiest assumption is the active-RIS power-consumption model: if real switching-and-control power does not scale with element count and holding-interval count the way the paper's Section II-B assumes, both the optimal holding interval and the claimed energy savings change.

Editorial extensions

If this is right

  • Operators can translate an energy budget directly into a maximum RIS switching frequency, because the outer-layer search returns the holding interval that just satisfies the energy constraint.
  • Blocked canyon and urban satellite links can be served with an active RIS that reconfigures only once per several seconds, avoiding the cost of per-slot RIS updates.
  • The joint direction optimization is worth its overhead: fixing the RIS direction or optimizing only the phase shifts causes a visible rate drop in the paper's simulations.
  • Active RIS with the proposed design achieves a target average rate with fewer elements than passive RIS and tolerates larger RIS-user distances, widening the usable deployment area.

Reading between the lines

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

  • The slow/medium/fast variable split is general enough that the same formulation should apply to any relay with a mechanically steered antenna and a fast electronic beamformer, not just active RIS.
  • A natural extension the paper does not pursue is to co-optimize the aerial RIS position and element count with the energy budget, since the distance sensitivity shown in the simulations depends on both.
  • An experimental realization with finite phase quantization and imperfect Doppler compensation would reveal how much of the simulated margin over passive RIS survives in hardware.
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. This paper considers an active RIS-assisted LEO satellite downlink in an NLoS canyon scenario, where the satellite-RIS link is blocked and the active RIS provides a virtual LoS path to the user. The authors propose a three-timescale optimization framework: satellite transmit beamforming is updated every unit time slot, active RIS beamforming is updated per holding interval, the RIS deployment direction is fixed for the whole communication period, and the holding interval length K is optimized in an outer loop. The inner-layer optimization uses FP, AO, SCA, and penalty methods to solve subproblems for the transmit beamformers, the RIS beamforming matrices, and the RIS orientation, subject to a unit-norm satellite beamformer, amplitude limits, and an active-RIS energy budget. Numerical results in Figs. 3-5 claim improvements in instantaneous and average achievable rate over self-generated SDR/GR and partial-optimization baselines, as well as reduced energy consumption through the choice of K.

Significance. If the equation-level problems are corrected, the three-timescale decomposition is a plausible and useful engineering contribution: it addresses a real trade-off between RIS beamforming adaptivity and switching energy in a fast-varying satellite channel, and it explicitly accounts for the RIS deployment direction as an optimization variable. The authors provide a complete algorithmic structure and a complexity analysis, and the simulation study covers rate, holding-interval duration, and energy-budget effects. However, the paper does not provide code or comparisons against published LEO-RIS benchmarks, and the numerical claims, especially the energy-saving and optimal-K claims, rest on an energy-constraint formulation that is currently inconsistent with the stated physical energy model. The paper is a letter, but the load-bearing algebraic and dimensional errors in Eqs. (3), (5), and (8) must be fixed before the conclusions can be accepted.

major comments (3)
  1. [Section III-B2, Eq. (5), constraint C4] The energy constraint in the RIS beamforming subproblem is algebraically inconsistent with the energy model in Section II-B. The model defines the first energy contribution as eta ||c1 sqrt(r^T l_sr) Theta_b H_sr w_{b,k}||^2, which, with V_b = theta_b theta_b^H, equals eta c1^2 r^T l_sr theta_b^H diag(|H_sr w_{b,k}|^2) theta_b. Problem (5), however, uses Tr(A3 V_b) with A3 = H_sr w w^H H_sr^H, which equals |theta_b^H H_sr w|^2. These two expressions coincide only if H_sr w has a single nonzero entry; in general they differ because Tr(A3 V_b) permits phase cancellations across RIS elements to reduce the reported energy without reducing the elementwise output power. Since Algorithm 1's outer loop stops when this C4 becomes infeasible, the selected holding interval K and the energy-savings claims in Figs. 4-5 are not grounded in the physical model. Please replace A3 by diag(|H_sr w_{b,k}|^2) (equivalently diag(H_sr w_{b,k}) diag(H_sr w_{b,k})^H) in (5), propagate the correction to (7), and re-run the simulations. I note that the corresponding term in problem (9) uses ||Theta_b H_sr w||^2 and is dimensionally consistent, so the conflation is specific to the V_b reformulation in (5)-(7).
  2. [Section III-B3, Eq. (8)] The definition c8 = c1 H_sr^H Theta_b H_sr w_{b,k} is dimensionally wrong. Since H_sr is an NM x L matrix and Theta_b is NM x NM, the product H_sr^H Theta_b H_sr is an L x L matrix, so c8 is an L x 1 vector, yet it is used as a scalar in the terms c8 sqrt(r^T l_sr r^T l_ru) and c8^2. The scalar effective channel should be c1 H_ru^H Theta_b H_sr w_{b,k}; the H_sr^H and H_ru^H factors appear to have been swapped. This error propagates into the definitions of c10, A4, and the orientation subproblem (9)-(10), making the orientation optimization as written undefined. Please correct c8 and verify dimensionality throughout the orientation subproblem.
  3. [Section III-B1, Eq. (3)] The matrix H1 is defined as (H_sr^H Theta_b H_ru)(H_ru^H Theta_b H_sr). Since Theta_b = diag({A_{m,n} e^{j phi_{m,n}}}) is not Hermitian, the quadratic form |H_ru^H Theta_b H_sr w|^2 expands to w^H H_sr^H Theta_b^H H_ru H_ru^H Theta_b H_sr w; the first factor should contain Theta_b^H rather than Theta_b. As written, H1 is generally non-Hermitian, so the stated Rayleigh-quotient solution, with w* = ell/||ell|| where ell is the eigenvector of the maximum eigenvalue, is not justified. This is load-bearing for the transmit beamforming update that is executed in every AO iteration.
minor comments (5)
  1. [Section III-C, Algorithm 1] The title 'Algorithm for solving Problem (6)' should refer to Problem (2) or to the inner-layer problem; additionally, the symbol R is overloaded between the set of rate values in step 10 and the rank-one matrix R in problem (9), which is confusing.
  2. [Section III-B3, Eq. (9)] The definition c10 = 2 mu_{b,k} sqrt(1 + mu_{b,k}) c8 appears to contain a typo: based on Eq. (4), it should be 2 mu_{b,k} sqrt(1 + v_{b,k}) c8, with v_{b,k} rather than mu_{b,k} inside the square root.
  3. [Section II-B] The definition c1 = sqrt(pi G_R G_U G_S P_S s_{b,k}) is dimensionally ambiguous; if s_{b,k} is the unit-power data symbol, it should not appear inside c1, and if P_S is the satellite transmit power, the notation should be P_S with a clear subscript to avoid confusion with the symbol for the communication period.
  4. [Section III-B2, Eqs. (5) and (9)] The constants c6, c7, and c10 depend on b and k but are written without subscripts, which makes the summation structure in (5) and (9) hard to follow; please add the b,k dependence explicitly.
  5. [Section III-C, complexity analysis] The complexity expression O(I_K I_a (2 I_f (...) + B K L^3) has an unbalanced parenthesis and uses both epsilon and the convergence tolerance ǫ; please reconcile the notation and close the parentheses.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity; the derivation chain is self-contained, with only a minor non-load-bearing self-citation to [12].

full rationale

The central derivation is self-contained rather than circular. Problem (2) maximizes an average achievable rate defined from the channel model in Section II-B, subject to an energy budget C4 taken from the external active-RIS model of [6] and a unit-norm orientation constraint; no parameter is fitted to the quantity being predicted, and the reported rate-versus-K tradeoff is a genuine consequence of fixing the RIS beamforming matrix over longer holding intervals. The inner-layer steps use standard FP/AO/SCA/penalty updates and do not reintroduce the objective as an input. The only self-citation is [12], a prior paper by co-author B. Lyu, used as a methodological reference for the penalty method and complexity counting; it is not load-bearing because the penalty approach is standard and the optimization would stand without it. The numerical section compares against self-generated SDR/GR and partial-optimization baselines, which is weak external validation but not circularity. A separate algebraic issue exists: Section II-B defines the amplification energy with ||Theta_b H_sr w||^2, whereas C4 in (5) uses Tr(A3 V_b) = |theta_b^H H_sr w|^2; these differ unless the RIS elements share a common phase. That is a correctness and consistency defect in the energy reformulation, not a circular reduction, so it does not raise the circularity score. Overall, no predicted quantity is equivalent by construction to an input, so the paper's derivation chain is not circular.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on imported channel, radiation, and energy models from prior work (Eq. 1 and Section II-B), plus standard convex optimization convergence assumptions. No parameters are fitted to external data, but several simulation parameters are hand-set and directly shape the numeric results. No new physical entities are introduced.

free parameters (5)
  • Radiation exponent beta = 1
    Set to 1 in Eq. (1) for simplicity; controls the shape of the RIS radiation pattern and directly enters the orientation subproblem and SNR.
  • Rician factor kappa = 3
    Hand-set in Section IV for the aRIS-user channel; fixes the LoS-to-scattered power ratio and influences all rate results.
  • Amplifier efficiency reciprocal eta = 1.25
    Set in Section IV; scales the amplification energy in constraint C4 and therefore determines which holding intervals K are feasible.
  • Maximum amplification a_max = 10 dB
    Set in Section IV as the upper bound in C5; limits the RIS amplification coefficients during beamforming optimization.
  • Energy budget E_max = not stated explicitly
    Constraint C4 and Fig. 5 use E_max, but the specific budget values used for the curves are not reported in the visible text.
assumptions (4)
  • domain assumption Normalized radiation pattern F_RIS(r,l) = (r^T l)^beta for 0 <= r^T l <= 1 and 0 otherwise, taken from [1],[2] (Eq. 1).
    The RIS orientation optimization and the SNR scaling both depend on this model; no experimental validation is provided in this paper.
  • domain assumption Active RIS energy model from [6]: amplifier power proportional to output signal power plus noise amplification, plus switching/control power B M N P_C.
    Used in Section II-B and constraint C4; the entire energy-saving claim rests on this model, which is imported without re-derivation.
  • domain assumption Channel and satellite position are constant within each 1 s unit time slot, and Doppler is perfectly compensated.
    Stated in Section II-B; justifies per-slot transmit beamforming and per-holding-interval RIS beamforming.
  • standard math Standard convergence properties of FP, SCA, and penalty methods yield a near-optimal feasible point for nonconvex problems (4), (7), and (10).
    Section III relies on these convergence results without proving them in the letter.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Active RIS Enabled NLoS LEO Satellite Communications: A Three-timescale Optimization Framework." pith.science (2026). https://pith.science/paper/AUNDF553

@misc{pith2026250620424,
  author       = {Pith},
  title        = {Pith review of: Active RIS Enabled NLoS LEO Satellite Communications: A Three-timescale Optimization Framework},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AUNDF553}},
  note         = {Machine review of arXiv:2506.20424}
}
read the original abstract

In this letter, we study an active reconfigurable intelligent surfaces (RIS) assisted Low Earth orbit (LEO) satellite communications under non-line-of-sight (NLoS) scenarios, where the active RIS is deployed to create visual line-of-sight links for reliable communication. To address the challenges of high energy consumption caused by frequent beamforming updates in active RIS, we propose a three-timescale optimization framework that jointly designs the transmit beamforming, RIS beamforming, and RIS direction vectors based on their characteristics. The goal is to maximize the system achievable rate while reducing energy consumption by controlling the RIS beamforming switching frequency. Then, a two-layer solution framework is developed, incorporating fractional programming (FP), alternating optimization (AO), successive approximation (SCA), and penalty-based methods, to obtain the optimized solution. Simulation results demonstrate that the proposed scheme can effectively improve system performance and reduce the energy consumption of the active RIS.

Figures

Figures reproduced from arXiv: 2506.20424 by the authors.

Figure 1
Figure 1. Active RIS-aided LEO Satellite communication syste [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Illustration of three-timescale framework. [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 5
Figure 5. Average achievable rate versus the aRIS-User distan [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Average achievable rate versus the holding interval [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages

  1. [1]

    RIS-Enhanced LEO Satellite Communicat ion: Joint Pas- sive Beamforming and Orientation Optimization,

    Z. Zheng et al., “RIS-Enhanced LEO Satellite Communicat ion: Joint Pas- sive Beamforming and Orientation Optimization,” 2022 IEEE Globecom W orkshops., Rio de Janeiro, Brazil, 2022, pp. 874-879

    work page 2022

  2. [2]

    Wireless Communications With Reconfigura ble Intelligent Surface: Path Loss Modeling and Experimental Measurement,

    W. Tang et al, “Wireless Communications With Reconfigura ble Intelligent Surface: Path Loss Modeling and Experimental Measurement, ” IEEE Trans. Wireless Commun ., vol. 20, no. 1, pp. 421-439, Jan. 2021

    work page 2021

  3. [3]

    Intelligent Reflecting Surface Op eration Under Predictable Receiver Mobility: A Continuous Time Propagat ion Model,

    B. Matthiesen et al., “Intelligent Reflecting Surface Op eration Under Predictable Receiver Mobility: A Continuous Time Propagat ion Model,” IEEE Wireless Commun. Lett ., vol. 10, no. 2, pp. 216-220, Feb. 2021

    work page 2021

  4. [4]

    Enabling NLoS LEO Satellite Communications with Reconfigurable Intelligent Surfaces

    X. Tian, N. Gonzalez-Prelcic, and T. Shimizu, “Enabling NLoS LEO Satellite Communications with Reconfigurable Intellig ent Surfaces,” arXiv:2205.15528, 2022

    work page Pith review arXiv 2022

  5. [5]

    RIS-Aided MIMO Downlink Transmission fo r Ultradense LEO Satellite-Terrestrial Networks,

    X. Zhang et al,“RIS-Aided MIMO Downlink Transmission fo r Ultradense LEO Satellite-Terrestrial Networks,” IEEE Internet Things J ., vol. 12, no. 11, pp. 15304-15318, June, 2025

    work page 2025

  6. [6]

    Active Re configurable Intelligent Surface-Aided Wireless Communications,

    R. Long, Y . -C. Liang, Y . Pei, and E. G. Larsson, “Active Re configurable Intelligent Surface-Aided Wireless Communications,” IEEE Transactions on Wireless Communications ., vol. 20, no. 8, pp. 4962-4975, Aug. 2021

    work page 2021

  7. [7]

    Active reconfigurable intelligent surfac e assisted secure and robust cooperative beamforming for cognitive satellit e-terrestrial net- works,

    Ge Y and Fan J. “Active reconfigurable intelligent surfac e assisted secure and robust cooperative beamforming for cognitive satellit e-terrestrial net- works,” IEEE Trans. V eh. Technol., vol. 72, no. 3, pp. 4108-4113, March 2023

    work page 2023

  8. [8]

    Intelligent reflecting surface enhan ced wireless net- works: Two-timescale beamforming optimization,

    M. -M. Zhao et al., “Intelligent reflecting surface enhan ced wireless net- works: Two-timescale beamforming optimization,” IEEE Trans. Wireless Commun, vol. 20, no. 1, pp. 2-17, Jan. 2021

    work page 2021

Show all 15 references
  1. [9]

    Fractional Programming for Communica tion Systems—Part II: Uplink Scheduling via Matching,

    K. Shen and W. Y u, “Fractional Programming for Communica tion Systems—Part II: Uplink Scheduling via Matching,” IEEE Trans. Signal Process., vol. 66, no. 10, pp. 2631-2644, May, 2018

    2018

  2. [10]

    Intelligent Reflecting Surface Enha nced Wireless Network via Joint Active and Passive Beamforming,

    Q. Wu and R. Zhang, “Intelligent Reflecting Surface Enha nced Wireless Network via Joint Active and Passive Beamforming,” IEEE Trans. Wireless Commun., vol. 18, no. 11, pp. 5394-5409, Nov. 2019

    2019

  3. [11]

    A Novel SCA-Based Method for Beamformin g Optimization in IRS/RIS-Assisted MU-MISO Downlink,

    V . Kumar et al. “A Novel SCA-Based Method for Beamformin g Optimization in IRS/RIS-Assisted MU-MISO Downlink,” IEEE Wireless Commun. Lett. , vol. 12, no. 2, pp. 297-301, Feb. 2023

    2023

  4. [12]

    Energy-Efficiency Maximization for STAR -RIS Enabled Cell-Free Symbiotic Radio Communications,

    B. Lyu at al., “Energy-Efficiency Maximization for STAR -RIS Enabled Cell-Free Symbiotic Radio Communications,” IEEE Trans. Cognit. Com- mun. Networking., vol. 10, no. 6, pp. 2209-2223, Dec. 2024

    2024

  5. [13]

    Energy Efficient Aerial RIS: Phase Shift Optimization and Trajectory Design,

    H. E. Hammouti et al. “Energy Efficient Aerial RIS: Phase Shift Optimization and Trajectory Design,” 2024 IEEE V eh. Technol. Conf. , Singapore, Singapore, 2024, pp. 1-7

    2024

  6. [14]

    Refracting RIS-aided hybrid satellite-t errestrial relay networks: Joint beamforming design and optimization,

    Lin Z et al., “Refracting RIS-aided hybrid satellite-t errestrial relay networks: Joint beamforming design and optimization,” IEEE Trans. Aerosp. Electron. Syst. , vol. 58, no. 4, pp. 3717-3724, Aug. 2022

    2022

  7. [15]

    Zhang, Matrix Analysis and Application

    X.-D. Zhang, Matrix Analysis and Application. , Cambridge, U.K.: Cambridge Univ. Press, 2017

    2017

Pith tools

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