{"id":"95a7398d-6301-4f40-8bed-a4b1950c2219","arxiv_id":"2606.10190","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"MA-RIS with SDE-modeled motion and overhead-aware two-timescale control reaches 36 dB steady-state SNR and 16x higher energy efficiency than active RIS in simulations.","lead":"This paper proposes a movable-antenna RIS system that physically repositions elements using stochastic motion models to escape deep fades, then optimizes movement and phases on two timescales. A generalist might read it to see whether adding physical mobility to smart surfaces can deliver more reliable, energy-efficient wireless links than static or active-RIS designs.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"SDE trajectory model and predictive HJB approximation remain the least-secured assumptions behind the reported 36 dB SNR and EE gains","rationale":"Reader's weakest_assumption directly identifies the same modeling gap. Full-text simulations do not close it; therefore the UNVERDICTED verdict and low confidence are retained. No other internal inconsistency or missing baseline rises to the same load-bearing level.","tokens_in":1802,"tokens_out":309,"duration_ms":11613,"concrete_test":"Re-run the steady-state distribution and SNR curves of §IV using an SDE whose diffusion term is increased by 2× (or replaced by a measured position histogram from a physical MA testbed); if the reported 36 dB mean SNR drops by >3 dB or the stability metric changes by >20 %, the modeling assumptions materially affect the central claim.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline numerical claims rest on the SDE (controlled drift + diffusion) producing realistic steady-state distributions via Ito calculus and on the predictive HJB approximation incurring negligible error in the two-timescale controller. The manuscript validates both only through Monte-Carlo simulations under the same generative model; no sensitivity analysis to diffusion coefficient mismatch, no discrete-time implementation error bounds, and no hardware trajectory traces are provided. If either modeling step deviates materially from physical antenna motion or real-time computation, the 15 dB / 30 dB / 16× gains do not transfer.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper proposes a Movable Antenna-enhanced RIS (MA-RIS) architecture in which antenna elements are physically repositioned according to a stochastic differential equation (SDE) model that includes controlled drift and environmental diffusion. Ito calculus is used to derive steady-state antenna distributions, spatial decorrelation, and outage probability. An overhead-aware two-timescale framework is introduced that separates slow trajectory optimization from fast phase adaptation; the resulting stochastic control problem is solved via a predictive approximation to the Hamilton-Jacobi-Bellman (HJB) equation. Monte-Carlo simulations are reported to confirm up to 36 dB steady-state SNR, gains of 15 dB over position-only control and 30 dB over uncontrolled baselines, and up to 16\times higher energy efficiency than active RIS.","tokens_in":1897,"tokens_out":596,"duration_ms":22294,"significance":"If the SDE trajectory model and predictive HJB approximation are shown to be robust, the work would establish mobility-induced diversity as a practical complement to conventional RIS phase control, offering a route to both higher SNR stability and substantially improved energy efficiency. The explicit treatment of control overhead via the two-timescale separation is a concrete engineering contribution.","major_comments":[{"comment":"Abstract and simulation-validation paragraph: the headline claims (36 dB SNR, 15 dB / 30 dB gains, 16\times EE) rest entirely on Monte-Carlo trials generated from the identical SDE model used for analysis; no sensitivity sweeps on the diffusion coefficient, no discrete-time discretization error bounds, and no hardware trajectory traces are supplied. Because these numbers are the central empirical support for the two-timescale controller, the absence of external validation is load-bearing.","section":"Abstract"},{"comment":"SDE / Ito-calculus analysis and predictive HJB sections: the steady-state distributions and real-time controller both rely on the SDE framework and the HJB approximation incurring negligible error, yet the manuscript provides neither analytic error bounds on the predictive approximation nor any comparison against an exact dynamic-programming solution or measured antenna motion. These modeling steps directly underpin the reported stability and EE advantage.","section":"Modeling and optimization framework"}],"minor_comments":[{"comment":"The abstract states that 'simulations validate theoretical predictions' but supplies neither the system parameters (number of elements, diffusion strength, carrier frequency) nor the precise definitions of the position-only and active-RIS baselines, impeding immediate reproducibility.","section":"Abstract"},{"comment":"Notation for the controlled SDE (drift term, diffusion matrix) and the two-timescale separation should be introduced with explicit equations early in the manuscript to aid readers who are not specialists in stochastic control.","section":"Introduction / System model"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the thorough and constructive review. The comments highlight important aspects of validation and modeling assumptions. Below we respond point-by-point, indicating where revisions have been made to the manuscript.","responses":[{"response":"We agree that all reported numerical results are generated from the proposed SDE model, which is standard practice when introducing a new analytical framework. In the revised manuscript we have added a new subsection in the simulation section that performs sensitivity sweeps over the diffusion coefficient, confirming that the reported SNR gains and EE improvements remain consistent across a wide range of diffusion values. We have also included a brief convergence argument showing that the discrete-time simulation error vanishes as the time step approaches zero, consistent with the underlying Itô calculus. Hardware trajectory traces are not supplied because the work is a theoretical and simulation-based study; we have updated the abstract and introduction to explicitly state that all results are obtained via Monte-Carlo simulation of the SDE model.","revision_made":"partial","referee_comment":"[Abstract] Abstract and simulation-validation paragraph: the headline claims (36 dB SNR, 15 dB / 30 dB gains, 16 times EE) rest entirely on Monte-Carlo trials generated from the identical SDE model used for analysis; no sensitivity sweeps on the diffusion coefficient, no discrete-time discretization error bounds, and no hardware trajectory traces are supplied. Because these numbers are the central empirical support for the two-timescale controller, the absence of external validation is load-bearing."},{"response":"The predictive HJB approximation is motivated by the two-timescale separation, under which the trajectory evolves slowly relative to phase adaptation. In the revision we have added an error-bound derivation in the appendix that quantifies the approximation error under the assumption of bounded diffusion; the bound is shown to be small for the parameter regimes considered. Exact dynamic programming is computationally intractable for the continuous-state problem, but we have included a numerical comparison against a discretized dynamic-programming solution in a simplified low-dimensional setting in the supplementary material, demonstrating close agreement. Measured antenna motion data are not available, as the study does not include hardware experiments; we have added an explicit limitations paragraph noting this scope.","revision_made":"partial","referee_comment":"[Modeling and optimization framework] SDE / Ito-calculus analysis and predictive HJB sections: the steady-state distributions and real-time controller both rely on the SDE framework and the HJB approximation incurring negligible error, yet the manuscript provides neither analytic error bounds on the predictive approximation nor any comparison against an exact dynamic-programming solution or measured antenna motion. These modeling steps directly underpin the reported stability and EE advantage."}],"tokens_in":1482,"tokens_out":592,"duration_ms":23818,"standing_objections":["Supplying hardware trajectory traces or measured antenna motion, as the manuscript is a purely theoretical and simulation-based study without experimental components."]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that this paper puts physical antenna motion into an RIS setup using an SDE for controlled drift plus diffusion, applies Ito calculus to get steady-state channel statistics, and then solves the long-term SNR problem with an overhead-aware two-timescale controller that approximates the HJB equation for real-time use.\n\nWhat is new is the specific combination: prior RIS papers treat positions as fixed or optimize them separately, while this one folds the stochastic trajectory directly into the control objective and separates slow movement from fast phase shifts. The simulations line up with the derived outage expressions and show the joint scheme reaching 36 dB SNR, beating position-only control by 15 dB and static baselines by over 30 dB, plus up to 16 times better energy efficiency than active RIS.\n\nThe soft spot is that all the numbers come from Monte Carlo runs generated by the same SDE the authors chose for design. There is no sensitivity check when the diffusion coefficient differs from reality, no discrete-time implementation bounds, and no hardware trajectory data. If the motion model is off, the reported gains shrink. The HJB predictive approximation is also only tested inside the model, so its real-time error remains unquantified.\n\nThis is aimed at researchers working on RIS, movable antennas, and stochastic wireless control. The framework is coherent and the math is laid out clearly enough that a serious referee can check the derivations and ask for the missing robustness tests. I would send it to peer review.","headline":"MA-RIS paper adds SDE motion modeling and two-timescale HJB control to RIS work, with solid simulated SNR and EE gains inside its own model but no external validation.","tokens_in":2429,"tokens_out":382,"would_cite":false,"duration_ms":12575,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Movable-antenna RIS with joint trajectory and phase optimization achieves 36 dB SNR and 16-fold energy efficiency gains","keywords":["movable antenna","reconfigurable intelligent surface","stochastic differential equation","two-timescale optimization","energy efficiency","signal-to-noise ratio","wireless propagation control"],"falsifier":"Physical experiments that track actual antenna paths and measure resulting SNR and power consumption, showing either steady-state SNR below the reported 36 dB or energy efficiency gains falling short of the 16-fold improvement over active RIS.","tokens_in":2661,"feed_emoji":"📡","tokens_out":697,"duration_ms":20466,"temperature":0.7,"pith_summary":"The paper establishes that adding physical mobility to RIS elements, modeled via stochastic differential equations with controlled drift and diffusion, creates spatial diversity that counters persistent deep fades in static wireless links. An overhead-aware two-timescale framework separates slow antenna repositioning from fast phase shifts and solves the resulting stochastic control problem through predictive Hamilton-Jacobi-Bellman approximation, enabling real-time operation while balancing control cost. Simulations confirm the approach yields stable high SNR and substantially better energy efficiency than active RIS, even though peak SNR is lower. A reader would care because this supplies a low-power route to resilient propagation control that exploits movement rather than amplification.","feed_headline":"Movable antennas on RIS reach 36 dB SNR at 16x efficiency","feed_subtitle":"Two-timescale position and phase control beats static and active RIS in stability and power use for wireless links.","key_machinery":"The overhead-aware Two-timescale framework that separates slow antenna trajectory control from fast phase adaptation, solved via predictive approximation of the Hamilton-Jacobi-Bellman equation.","core_discovery":"The authors introduce a Movable Antenna-enhanced RIS architecture in which antenna elements follow an SDE trajectory that combines deterministic control with environmental randomness. Ito-calculus analysis yields the steady-state distribution, spatial decorrelation, and outage probability. The long-term SNR maximization problem, incorporating control overhead, is addressed by a two-timescale strategy that optimizes slow trajectories via predictive HJB approximation while adapting phases on a fast scale, producing up to 36 dB steady-state SNR and up to 16 times higher energy efficiency than active RIS.","pith_inferences":["The SDE mobility model could be transferred to optimize trajectories in other dynamic-antenna or drone-relay systems.","Integration with learned predictors might reduce the need for precise real-time HJB solving in rapidly changing environments.","Lower hardware cost and power draw suggest the architecture could scale to dense outdoor deployments where active RIS is impractical."],"forward_implications":["The two-timescale strategy reaches up to 36 dB steady-state SNR with high stability.","It outperforms position-only control by up to 15 dB and uncontrolled baselines by over 30 dB.","Energy efficiency is up to 16 times higher than Active RIS across varying system scales.","Fundamental trade-offs appear between control strength and mobility randomness."],"fun_headline_variants":["MA-RIS movable antennas achieve 36 dB SNR via two-timescale framework","36 dB SNR achieved by MA-RIS with movable antennas and phase control","MA-RIS SDE model enables 36 dB SNR and 16x energy efficiency","Joint movement and phase optimization yields 36 dB SNR in MA-RIS"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The SDE model together with its Ito-calculus steady-state analysis faithfully represents actual antenna trajectories and diffusion, and the predictive HJB approximation incurs negligible error under real-time constraints.","fun_headline_variants_meta":{"raw":{"variants":["MA-RIS movable antennas achieve 36 dB SNR via two-timescale framework","36 dB SNR achieved by MA-RIS with movable antennas and phase control","MA-RIS SDE model enables 36 dB SNR and 16x energy efficiency","Joint movement and phase optimization yields 36 dB SNR in MA-RIS"]},"model":"grok-4.3","cost_usd":0.004733,"raw_usage":{"total_tokens":2277,"prompt_tokens":714,"num_sources_used":0,"completion_tokens":85,"cost_in_usd_ticks":47328000,"prompt_tokens_details":{"text_tokens":714,"audio_tokens":0,"image_tokens":0,"cached_tokens":64},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1478,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":714,"tokens_out":85,"duration_ms":10457,"temperature":1.0,"reasoning_tokens":1478,"cache_read_input_tokens":64,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-27T15:09:35.888554+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Physical experiments that track actual antenna paths and measure resulting SNR and power consumption, showing either steady-state SNR below the reported 36 dB or energy efficiency gains falling short of the 16-fold improvement over active RIS.","supporting_citations":[],"review_version":1}