{"id":"5d53daf3-c895-4b32-b329-a3d2ec6d9e46","arxiv_id":"2501.02256","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"An acoustic RIS placed at the sound-channel axis in deep water or near the seabed in shallow water can redirect sound into shadow zones, raising simulated coverage from under 20% to near 100%.","lead":"This paper proposes using acoustic reconfigurable intelligent surfaces (aRIS) to bounce underwater sound into shadow zones where nodes normally lose connectivity. Simulations indicate that carefully placed aRIS devices can lift coverage from under 20% to near 100%, and a small pool test shows the reflecting hardware works.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The near-100% coverage claim is not yet established because the deep-sea Bellhop validation models the aRIS as an independent source, which is the same assumption Theorem 1 relies on; a real aRIS must first capture and amplify energy from the primary source.","rationale":"The reader's weakest assumption already identifies source equivalence, and my independent read agrees that it is the single most load-bearing step because the abstract's headline numbers are produced by Bellhop runs whose aRIS appears to be a replacement source. The pool test gives real but narrow evidence (two elements, zero-degree reflection, short range) and does not close this gap. I also note a second, independent weakness in Theorem 1: maximizing only rmax is not shown to maximize rmax minus rmin, so even a hypothetical ideal-source aRIS is not rigorously proved optimal. I do not see grounds to reject the paper outright; the concept is plausible and the hardware work is a genuine contribution, but the coverage claim should be conditional on a power-budget-respecting simulation or link-budget analysis. Since the reader already assigned CONDITIONAL for essentially this reason, my recommendation is UNCHANGED.","tokens_in":12487,"tokens_out":3914,"duration_ms":42764,"concrete_test":"Re-run the deep-sea case of Fig. 9 with a single physical source at 200 m and model each aRIS as an N by N array of transducers with the measured element gain (2.69 dB for two units, scaled as 10 log10(N^2) for larger arrays) and a phase profile, rather than as an independent point source. As a minimal analytical version, compute the transmission loss from the 200 m source to the proposed 2100 m aRIS location in Bellhop, then compare the resulting received intensity at the aRIS with the intensity that the current simulation implicitly assigns to the 'aRIS source.' If the received intensity is below that level by more than the available array gain, the near-100% coverage figure is not supported. Run the same comparison for at least one non-optimal depth to see whether the optimality conclusion survives.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section IV-A opens the proof of Theorem 1 with \"Here, we equate an aRIS to a source,\" and the deep-sea Bellhop validation in Section V-A appears to follow the same substitution: multiple sources placed at the sound-channel axis are said to \"redirect energy into shadow zones.\" The central claim of near-100% coverage therefore depends on a physical premise that is never checked: that a finite-gain aRIS at the sound-channel axis receives enough incident energy from the 200 m source to behave like the source used in the simulation. This is not a minor implementation detail. The proposed \"first absorb, then radiate\" design contains amplifiers and phase shifters, so its re-radiated level is bounded by the incident field at the array times the array gain; no array gain can recover energy that was never there. The pool test demonstrates only that two elements can steer a reflected beam at short range; it does not quantify long-range incident capture. Separately, even under the source-equivalence assumption, Theorem 1 maximizes only rmax, the span of the surface-tangent ray, while the coverage area in Eq. (3) is determined by rmax minus rmin; the proof does not show that the depth maximizing rmax also maximizes the area, and the statement \"increasing r effectively reduces the shadow zone area\" is asserted rather than derived. Both issues are load-bearing because they are what convert a geometrical observation about where sources perform well into the paper's engineering claim that aRIS deployment covers shadow zones.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes using acoustic Reconfigurable Intelligent Surfaces (aRIS) to actively fill underwater acoustic shadow zones, which are regions of severe signal loss caused by sound-speed refraction. The authors analytically model shadow zones in deep-sea and shallow-sea environments, prove a deep-sea optimal deployment rule (sound channel axis), formulate a shallow-sea placement optimization, redesign a piezoelectric-based aRIS hardware with a 'first absorb, then radiate' unit, validate the hardware in a pool test, and use Bellhop ray-tracing simulations to claim that without aRIS coverage is below 20% while with optimal aRIS deployment coverage approaches nearly 100%.","tokens_in":12789,"tokens_out":5436,"duration_ms":53011,"significance":"If the coverage claims are validated, the work would represent a meaningful advance over the current passive shadow-zone-avoidance paradigm in underwater acoustic networking, with a concrete hardware prototype and a plausible deployment strategy. The pool-test measurement of a two-element beamforming gain near the theoretical 3 dB is a useful practical data point. However, the central performance claim rests on an ideal-source assumption that is shared by both the analytical proof and the numerical validation; the physical energy-capture problem is not addressed. The paper is therefore potentially significant but requires more rigorous proof and a physically grounded simulation before its headline results can be accepted.","major_comments":[{"comment":"The proof equates the aRIS to a source and then compares only the horizontal span rmax of the surface-tangent ray (Eqs. (15)-(17)). However, the coverage area defined in Eq. (3) is determined at each depth by the difference rmax(z) - rmin(z), and the total coverage area in Eq. (6) integrates this difference. The statement in the proof that 'a larger V-shaped span implies a greater rmax - rmin at each depth' is asserted without derivation. As written, Theorem 1 proves only that the sound channel axis maximizes a single ray's horizontal span, not that it maximizes the full coverage area. Please provide a proof that the depth maximizing the boundary span also maximizes the integral of rmax(z) - rmin(z) over the region of interest, or revise the theorem statement to match what is actually established.","section":"§IV-A, Theorem 1"},{"comment":"The Bellhop validation appears to model the aRIS as an independent source placed at the sound channel axis, which is exactly the same assumption used in the proof of Theorem 1. This does not independently validate the physical premise that a finite-gain aRIS, which must first absorb and amplify the incident field from the 200-m source, can reradiate enough energy at 2100-m depth to behave like the source in the simulation. The pool test in §III-B demonstrates only short-range beam steering with two elements and does not measure end-to-end capture-and-reradiate gain or the achievable source level at long range. Please add an explicit link-budget analysis (incident transmission loss at the aRIS, array gain for N elements, amplifier power constraints) and, if possible, a Bellhop configuration in which the aRIS is modeled as a receiver-amplifier-radiator with a finite gain derived from this budget, rather than as an independent source. Without this, the 'nearly 100%' coverage claim is conditional on an unvalidated assumption.","section":"§V-A and §IV-A"},{"comment":"The abstract and conclusions state that optimal aRIS deployment 'achieves nearly 100% energy coverage,' but the left panel of Fig. 10 indicates that this occurs only at the highest transmission-loss threshold considered (150 dB). At lower thresholds, the covered proportion is evidently smaller. Please report the coverage proportion at each threshold and state the corresponding source level and range assumptions, or explicitly qualify the 'nearly 100%' claim as applying only to the highest threshold simulated. The current wording overgeneralizes the simulation result.","section":"§V-A, Fig. 10"}],"minor_comments":[{"comment":"The sentence 'which proves (13)' should refer to (12), the theorem's coverage gain definition, rather than to (13), which defines the coverage efficiency η.","section":"§IV-A, after Eq. (17)"},{"comment":"The opening sentence 'Sec. V-A models these impacts' appears to refer to Section VI-A, not Section V-A; please correct the cross-reference.","section":"§VI-A"},{"comment":"The 'first absorb, then radiate' unit is described as containing amplifiers and phase shifters, yet the text contrasts it with amplify-and-forward relays by saying it 'reflects incident waves immediately through the intrinsic piezoelectric effect without needing signal receiving, processing, and re-transmitting modules.' This is internally confusing, since absorption, amplification, and phase-shifting constitute a receive-process-transmit chain. Please clarify the intended distinction from an AF relay, and specify the power source for the amplifiers.","section":"§III-A"},{"comment":"The simulation setup (frequency, source level, number of Bellhop rays, array size of the aRIS, and the threshold for 'coverage') is only partially specified. Please provide a table of the simulation parameters so that the results are reproducible.","section":"§V-A"},{"comment":"The calculations of the 'required number of aRIS units' (e.g., approximately 10 units for 10 km in shallow water) are described only qualitatively. Please include the formula or algorithm used to obtain these numbers.","section":"§V-A and §V-B"},{"comment":"The claim of 'first time in the literature' for active shadow zone coverage is strong; please provide a more nuanced comparison with prior work on underwater relays and node placement to support this novelty statement.","section":"§I"}],"recommendation":"major_revision","confidential_remarks":"The paper tackles an important and timely problem, and the hardware prototype is a tangible contribution. The main concern is that the analytical optimality proof and the Bellhop validation both rely on the same ideal-source assumption, so the headline near-100% coverage claim is not yet independently supported. The authors should be asked to add a physical power-budget model or to substantially soften the claims. With that addition, the paper could become publishable; without it, the central quantitative claim remains unverified."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a genuinely new application of acoustic RIS to a real problem, and the deployment insight (deep-sea at channel axis, shallow at seabed) is plausible. But the headline \"almost 100% coverage\" claim is not yet backed up. The proof and the Bellhop validation both treat the aRIS as an independent source, and a real finite-gain array has to capture incident energy first. That missing link is load-bearing.\n\nWhat's new and good: the idea of actively filling shadow zones rather than routing around them is the right instinct, and the analytical shadow-zone model based on ray acoustics is standard and clean. The two-element \"first absorb, then radiate\" hardware design is a sensible step toward practicality, and the pool test, small as it is, does show controlled reflection with roughly the expected 3 dB beamforming gain. The dynamic-environment robustness analysis is a nice extra.\n\nSoft spots, in order of severity. First, Theorem 1 equates an aRIS to a source and then argues that maximizing rmax maximizes coverage area; but coverage area depends on rmax - rmin, and no proof shows the same depth maximizes both. The statement \"increasing r effectively reduces the shadow zone area\" is asserted, not derived. Second, the Bellhop validation appears to use the same source-equivalence: placing aRIS at the channel axis in the simulator is modeled as a source, so it does not independently test whether the aRIS can actually capture enough incident energy from the 200 m source to re-radiate at the needed level. The pool test only proves short-range beam steering, not long-range capture. Third, the hardware description contradicts itself: it says signals are \"processed through amplifiers and phase shifters\" and then says the aRIS \"reflects incident waves immediately... without needing signal receiving, processing, and re-transmitting modules.\" That cannot both be true. Also, the \"less than 20%\" versus \"almost 100%\" gap is shown at a particular TL threshold; the abstract makes it sound universal.\n\nWho benefits: researchers working on underwater acoustic networks or RIS-aided communications. The paper is a useful pointer for a novel deployment concept. It deserves a serious referee because the problem is important and the idea is new, but the referee should push for either a physical model that includes incident-field-dependent gain or a simulation that explicitly models the aRIS as a finite-gain reflector, plus a fix to the proof and the hardware text. As it stands, the strong coverage claim is plausible but unproven.","headline":"A new and plausible use of acoustic RIS for shadow-zone coverage, but the near-100% coverage claim rests on equating the RIS to an independent source, which is exactly what needs proving.","tokens_in":13288,"tokens_out":2626,"would_cite":false,"duration_ms":24058,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Acoustic smart surfaces can fill underwater shadow zones, lifting coverage from under 20% to nearly 100%.","keywords":["underwater acoustic communication","shadow zone","acoustic reconfigurable intelligent surface","sound speed profile","ray acoustics","beamforming","coverage optimization","deep-sea and shallow-sea networks"],"falsifier":"Measure, in Bellhop or at sea, the received energy in a shadow zone for aRIS deployments at several depths (including the axis) using a realistic finite-aperture aRIS model that respects the actual incident field from the primary source; if an off-axis depth yields equal or larger coverage area than the axis under that model, the theorem's deployment rule fails. A simpler check: compute the coverage sets from the full acoustic field (not the point-source equivalence) and compare the $r_{\\max}$ at each depth.","tokens_in":12284,"feed_emoji":"🌊","tokens_out":3589,"duration_ms":34053,"temperature":0.7,"pith_summary":"This paper argues that underwater acoustic shadow zones—regions tens of kilometres wide where sound bends away and no signal arrives—can be actively filled rather than passively avoided. The proposed tool is an acoustic Reconfigurable Intelligent Surface (aRIS): a panel of piezoelectric elements that absorbs an incoming sound wave and re-radiates it with a controlled phase, steering energy into the shadow zone. The paper develops a ray-acoustics model of shadow zones, proves a deep-sea deployment rule (place the aRIS at the sound-channel axis) and a shallow-sea rule (place it as deep as possible), supplies a hardware design validated in pool tests, and reports Bellhop simulations in which coverage rises from under 20% without the aRIS to nearly 100% with it. The significance, if the claims hold, is that a single controllable reflector could provide seamless connectivity for underwater networks in dynamic oceans.","feed_headline":"Acoustic RIS lifts underwater coverage from 20% to near 100%","feed_subtitle":"A reconfigurable acoustic mirror placed at the sound-channel axis can redirect energy into dead zones that no increase in source power can…","key_machinery":"The load-bearing object is the geometric shadow-zone model built from ray acoustics: with a depth-dependent sound speed $c(z)$, each ray's horizontal range $r(z)$ is an integral over grazing angle, and the coverage set is the region between the envelopes $r_{\\min}(z)$ and $r_{\\max}(z)$. In deep water this yields the V-shaped convergence-zone pattern; the aRIS is then equated to a point source positioned at its deployment depth, and Theorem 1 compares the maximum horizontal span $r_{\\max}$ of rays that graze the sea surface. The second machinery element is the aRIS unit itself: two piezoelectric elements in a 'first absorb, then radiate' configuration that together allow phase-controlled reflection and high-gain beamforming, which the paper validates with a two-element pool test showing roughly 3 dB of gain.","core_discovery":"The central claim is that an underwater region whose geometry makes it unreachable to direct rays can be covered by placing an acoustic RIS so that it re-radiates incident energy into that region, and that in a deep-sea sound channel the unique optimal depth for this re-radiation is the sound-channel axis, where sound speed is minimal. The paper expresses this as Theorem 1: comparing the horizontal span of the surface-tangent ray emitted from the aRIS depth, the span is largest when the aRIS sits at the axis, and since the coverage area is monotone in that span, the axis depth maximises coverage. For shallow seas, the paper treats rays as circular arcs under a linear sound-speed gradient and formulates an optimisation for the aRIS depth, concluding that the seabed position yields the longest coverage distance because it creates the largest reflection grazing angles.","pith_inferences":["If the source-equivalence assumption at the heart of Theorem 1 is relaxed, the optimal depth could shift: at the axis the incident intensity from a distant source is lower than at shallower depths, so a real aRIS may need more surface area than the paper's unit-count analysis suggests.","The coverage metric used (fraction of grid points above a transmission-loss threshold) does not directly measure bit error rate or data rate; a testable extension would quantify whether the redirected energy actually supports demodulation in the shadow zone.","The claim that energy coverage reaches 'almost 100%' is demonstrated for a single environmental snapshot; a natural extension is to sweep seasonal sound-speed-profile variations and count how often the fixed deployment rule remains near-optimal."],"forward_implications":["If the deep-sea rule is correct, a single pair of aRIS units placed symmetrically at the sound-channel axis can eliminate the first shadow zone, something no increase in source level accomplishes.","In shallow seas, placing the aRIS at the seabed roughly halves the number of units needed to cover a 10 km shadow zone compared with non-optimal depths (about 10 vs 20 units).","The viability of the two-element 'absorb-then-radiate' unit in pool tests suggests the aRIS does not need simultaneous send/receive piezoelectric operation, lowering the hardware barrier.","Dynamic-platform compensation (phase correction for displacement and rotation) is claimed to restore coverage with ~99% RMSE reduction for translation and ~80% for rotation, supporting robustness claims."],"supporting_citations":[{"why":"Introduces the acoustic reconfigurable intelligent surface concept that this paper builds on for active shadow-zone coverage.","marker":"[13]"},{"why":"Provides the Munk deep-sea sound speed profile model used in the theoretical analysis and Bellhop simulations.","marker":"[19]"},{"why":"Supplies the Bellhop ray-tracing code used for all coverage and energy-distribution simulations.","marker":"[29]"},{"why":"Details acoustic RIS beamforming design, grounding the reflection and re-radiation machinery.","marker":"[15]"},{"why":"Defines the ideal RIS model that the practical two-element hardware design must approximate.","marker":"[22]"},{"why":"Supplies the sequential quadratic programming method used to solve the shallow-sea deployment optimisation.","marker":"[27]"},{"why":"Provides the circular-arc approximation of sound rays under a linear gradient, underlying the shallow-sea coverage-distance formula.","marker":"[25]"}],"fun_headline_variants":["Acoustic smart surfaces erase underwater dead zones","Underwater coverage jumps from 20% to 100% with acoustic RIS","Reconfigurable acoustic surfaces bridge ocean shadow zones","Acoustic RIS fills kilometers-long underwater gaps"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof that the sound-channel axis is the best depth treats the aRIS as if it were an ideal point source that receives enough energy from the primary source to re-radiate with the same wavefront, and it identifies maximum coverage with the single widest surface-tangent ray span.","fun_headline_variants_meta":{"raw":{"variants":["Acoustic smart surfaces erase underwater dead zones","Underwater coverage jumps from 20% to 100% with acoustic RIS","Reconfigurable acoustic surfaces bridge ocean shadow zones","Acoustic RIS fills kilometers-long underwater gaps"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000193,"raw_usage":{"total_tokens":1321,"prompt_tokens":886,"completion_tokens":435,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":502,"completion_tokens_details":{"reasoning_tokens":372}},"tokens_in":502,"tokens_out":435,"duration_ms":4213,"temperature":1.0,"reasoning_tokens":372,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:13:56.863744+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure, in Bellhop or at sea, the received energy in a shadow zone for aRIS deployments at several depths (including the axis) using a realistic finite-aperture aRIS model that respects the actual incident field from the primary source; if an off-axis depth yields equal or larger coverage area than the axis under that model, the theorem's deployment rule fails. A simpler check: compute the coverage sets from the full acoustic field (not the point-source equivalence) and compare the $r_{\\max}$ at each depth.","supporting_citations":[{"cited_title":"High-data-rate long-range underwater communications via acoustic reconfigurable intelligent surfaces,","cited_arxiv_id":null,"evidence_quote":"Introduces the acoustic reconfigurable intelligent surface concept that this paper builds on for active shadow-zone coverage."},{"cited_title":"Ocean acoustic tomography: Rays and modes,","cited_arxiv_id":null,"evidence_quote":"Provides the Munk deep-sea sound speed profile model used in the theoretical analysis and Bellhop simulations."},{"cited_title":"Bellhop code,","cited_arxiv_id":null,"evidence_quote":"Supplies the Bellhop ray-tracing code used for all coverage and energy-distribution simulations."},{"cited_title":"Designing acoustic reconfigurable intelligent surface for underwater communications,","cited_arxiv_id":null,"evidence_quote":"Details acoustic RIS beamforming design, grounding the reflection and re-radiation machinery."},{"cited_title":"Acoustic intelligent surface system for reliable and efficient underwater communications,","cited_arxiv_id":null,"evidence_quote":"Defines the ideal RIS model that the practical two-element hardware design must approximate."},{"cited_title":"Sequential quadratic programming,","cited_arxiv_id":null,"evidence_quote":"Supplies the sequential quadratic programming method used to solve the shallow-sea deployment optimisation."},{"cited_title":"The physical acoustics of underwater sound communication,","cited_arxiv_id":null,"evidence_quote":"Provides the circular-arc approximation of sound rays under a linear gradient, underlying the shallow-sea coverage-distance formula."}],"review_version":1}