{"id":"7e80ae5c-f8d7-4192-9eb5-4e0c1559876d","arxiv_id":"2607.07278","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":3,"one_line_summary":"Partitioning a transmit array into sub-arrays generating multiple coordinated Airy beams reduces sensitivity to obstacle-geometry estimation errors compared to single-Airy beamforming.","lead":"This paper proposes splitting a transmit antenna array into sub-arrays, each generating a different curved Airy beam, so that if one beam is blocked, others still reach the receiver. It matters because high-frequency 6G links are extremely vulnerable to obstacles, and this approach adds robustness without new hardware.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"Robustness bound (Eq. 58) is an upper bound on sensitivity, not a lower bound on performance; the constructive procedure leaves trajectory selection unspecified, so the claimed robustness rests on an unproven assumption that good diverse trajectories exist.","rationale":"The reader correctly identified the planar-screen model as a simplification, but this is a standard modeling choice in diffraction-based beam analysis and is acknowledged by the authors. The more load-bearing concern is internal: the theoretical robustness bound (Eq. 58) is an upper bound on sensitivity, not a constructive guarantee of robustness. The actual performance depends on trajectory selection, which is treated as an external black box. This means the paper proves 'multi-Airy *can* be robust' (there exist good trajectories) but does not prove 'multi-Airy *will* be robust' (a practical system will find them). That said, the simulation evidence is internally consistent and the phase-alignment derivation (Theorem 1) is correct. The contribution is real but incomplete — the reader's CONDITIONAL verdict with MODERATE confidence is appropriate. The concern I raise does not change the verdict because it is a gap in completeness rather than a correctness error: the simulations do demonstrate the claimed robustness for the selected parameters, and the authors honestly state that trajectory search is a separate stage. A stronger paper would either integrate trajectory selection into the optimization (43) or prove that the training procedure in [29] reliably finds diverse trajectories, but the absence of this does not invalidate the core contribution.","tokens_in":19360,"tokens_out":768,"duration_ms":344206,"concrete_test":"For the double-Airy case (M=2) with the simulation parameters in Section VI-A, compute χ̄_M from Eq. (58) for (a) the hand-selected diverse trajectories used in Fig. 9 and (b) a degenerate case where both sub-arrays use identical trajectory parameters (B_1=B_2, F_1=F_2, θ_1=θ_2). If χ̄_M is similar in both cases, the bound does not actually capture trajectory diversity, and the robustness claim relies entirely on trajectory selection rather than the multi-Airy structure itself.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is that multi-Airy beamforming reduces sensitivity to geometry estimation errors via trajectory diversity. The theoretical support is Eq. (58)-(59), which bound the relative energy variation χ_M. However, this bound is an *upper bound on the worst-case sensitivity* — it says the variation *cannot exceed* 2χ̄_M + χ̄_M². It does not guarantee that the multi-Airy energy J_M(x_e) is actually high; it only limits how much it can drop. The bound could be satisfied trivially by a configuration with near-zero nominal energy. The actual performance depends entirely on whether the trajectory parameters η = {(B_m, F_m, θ_m)} are well-chosen, but Algorithm 1 treats these as *specified inputs* and the trajectory-selection stage is explicitly deferred ('received-signal-power-based Airy beam training, such as the method in [29]'). This creates a gap: the robustness claim is demonstrated only for hand-selected trajectories in simulation, with no guarantee that a practical training procedure would find comparably diverse trajectories. If the beam-training stage converges to trajectories that overlap at the obstacle plane (which the paper itself flags as a failure mode: 'if the trajectories overlap strongly at the obstacle plane'), the multi-Airy advantage vanishes. The bound in Eq. (58) does not prevent this because the triangle inequality used to derive it is loose when trajectories are correlated.","agreement_with_reader":"partial"},"referee_report":{"model":"glm-5.2","summary":"This paper proposes a multi-Airy beamforming scheme for blockage-robust near-field communications. The authors first analyze the sensitivity of conventional single-Airy beamforming to transmitter-obstacle-receiver (Tx-obs-Rx) geometry estimation errors, showing that a small error can destroy the single dominant Airy main lobe. They then derive a phase-alignment rule (Theorem 1) for coherently combining multiple Airy beams at a target user, partition the transmit array into sub-arrays each generating a tailored Airy beam, and analyze a robustness bound showing that trajectory diversity reduces sensitivity. Simulations demonstrate significant rate improvements and robustness gains over single-Airy and focused beam schemes under a planar-screen blockage model.","tokens_in":20176,"tokens_out":955,"duration_ms":168524,"significance":"The paper addresses a practically important problem: blockage in high-frequency near-field links. The phase-alignment derivation (Theorem 1 and Appendix A) is clean and correct. The sensitivity analysis of single-Airy beamforming (Section IV-A, quantified by the 95.2% energy loss example) is a valuable contribution that motivates the multi-Airy approach. The robustness bound in Eq. (55)-(59), while an upper bound, provides useful intuition for why trajectory diversity helps. The UPA validation (Section VI-E) demonstrates extensibility beyond 1D arrays. Reproducible code is promised. The overall framework is constructive and the simulation results are convincing within the adopted model.","major_comments":[{"comment":"Section V-A, Eq. (43)-(47) and Algorithm 1: The trajectory parameters η = {(B_m, F_m, θ_m)} are treated as specified inputs, and the trajectory-selection stage is explicitly deferred to external beam training (e.g., [29]). This creates a gap between the theoretical robustness claim and the practical procedure. The robustness bound in Eq. (58) is an upper bound on sensitivity that does not guarantee high nominal energy J_M(x_e); it could be satisfied trivially by a configuration with near-zero energy. The paper itself flags trajectory overlap at the obstacle plane as a failure mode, but the bound does not prevent this because the triangle inequality is loose when trajectories are correlated. The authors should either (a) provide a trajectory-selection heuristic or criterion that explicitly promotes spatial diversity at the obstacle plane, or (b) clearly state as a limitation that therobst","section":null}],"minor_comments":[{"comment":"Section II-B, Eq. (6): The definition of b_l allows values > 1 as a geometric extrapolation, but this is only explained later. A forward reference or brief note at first introduction would help the reader.","section":null},{"comment":"Section III-B, Eq. (18): The Gaussian aperture width ω_0 and Airy truncation factor α_A are free parameters. Their values are not clearly specified in the simulation setup (Section VI-A). Please state the values used.","section":null},{"comment":"Section IV-A: The 95.2% energy loss is computed from a dB difference (-5.8 dB to -19.0 dB). It would help to show the linear-scale calculation explicitly for clarity.","section":null},{"comment":"Section VI-C, Fig. 9: The figure caption should specify which multi-Airy configuration is 'best' and list the trajectory parameters or at least the selection rule, so the comparison is reproducible.","section":null},{"comment":"Section VI-D, Fig. 10: At b_l = 0 (no blockage), the focused uniform beam appears to outperform some Airy schemes. Please discuss whether this is expected and whether Airy schemes incur a baseline penalty in unblocked scenarios.","section":null},{"comment":"Reference [14] (Darsena et al., arXiv:2508.13714) and several other arXiv preprints ([6], [15], [25], [29]) have 2025-2026 dates. Please verify these are correctly cited and update with published versions if available.","section":null},{"comment":"The notation eF in Eq. (21) uses a subscript that may render ambiguously. Please ensure it is clearly typeset as a single symbol 'e'. Please clarify.","section":null}],"recommendation":"minor_revision","confidential_remarks":"The skeptic's concern about the gap between the upper bound and actual performance is valid but does not undermine the central claim, which is supported by constructive simulation evidence. The paper is honest about the trajectory-selection deferral. I recommend minor revision with a request to address this gap more explicitly, either through a diversity-promoting selection criterion or a clearer limitation statement. The planar-screen model simplification is acceptable for a first analysis but should be acknowledged as a limitation."},"author_rebuttal":null,"desk_editor":{"model":"glm-5.2","letter":"The main thing to know: this paper proposes partitioning a transmit array into sub-arrays, each generating a different Airy beam trajectory, with derived phase offsets for coherent combining at the target. The phase-alignment derivation (Theorem 1, Eq. 29) is clean and correct — it's essentially a coherent-combining condition derived from first principles, and the finite-window refinement via cyclic coordinate ascent is a sensible practical extension. The sensitivity analysis (Eq. 55, 58) bounding target-window energy variation under geometry errors is also new and gives a useful qualitative explanation for why trajectory diversity helps. These are real contributions to the near-field Airy beamforming literature, and the simulation results showing 2.57× rate improvement over single-Airy under strong blockage are compelling within the model used. The authors also promise to release simulation code, which is good practice. The stress-test concern about Eq. 58 being an upper bound on sensitivity rather than a lower bound on performance is technically correct but somewhat overblown — the paper never claims the bound guarantees high nominal energy, and the authors explicitly acknowledge the failure modes (all trajectories blocked, trajectories overlapping at the obstacle plane). The bound is used as a qualitative explanatory tool, not as a performance certificate. That said, there is a genuine gap: Algorithm 1 treats trajectory parameters as specified inputs and defers trajectory selection to a separate beam-training stage. The robustness claim is demonstrated only for hand-selected trajectories in simulation. Whether a practical training procedure would find comparably diverse trajectories is unaddressed. This is the softest spot — not because the constructive procedure is wrong, but because the end-to-end story has a missing chapter. The planar-screen blockage model is simplified but standard for this type of analysis, and the UPA validation adds some generality. Some parameters (ω_0, α_A) are unspecified, which is a minor reproducibility issue. Overall: the paper does what it claims within its model. The phase-alignment method is the core contribution and it holds up. The robustness analysis is explanatory rather than prescriptive, but that's fine for what it is. Worth a serious referee who can push on the trajectory-selection gap and parameter reporting.","headline":"Multi-Airy beamforming with clean phase-alignment derivation; robustness bound is an upper bound on sensitivity, not a performance guarantee, and trajectory selection is a black box.","tokens_in":20261,"tokens_out":529,"would_cite":true,"duration_ms":85225,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["42.25.Bs","84.40.Ua"],"model":"glm-5.2","headline":"Multiple curved beams outperform single beam for obstacle blockage robustness","keywords":[],"falsifier":"The multi-Airy robustness advantage would collapse if the obstacle simultaneously blocks all sub-array trajectories, if the trajectories overlap at the obstacle plane (so a single error strip intersects multiple main lobes), or if sub-array partitioning reduces the nominal received energy J_M excessively. The paper states these conditions explicitly (Section V-B).","tokens_in":19601,"feed_emoji":"📡","tokens_out":886,"duration_ms":150559,"temperature":0.7,"pith_summary":"This paper addresses a specific vulnerability in high-frequency wireless communications: when using Airy beams—waves that follow curved trajectories to bend around obstacles—the beam's useful energy concentrates along a single path, so even a small error in estimating where the obstacle sits can cause the beam to miss its target entirely. The authors quantify this sensitivity, showing that a 9.4% geometry estimation error can destroy 95.2% of received signal energy. Their solution is to split the transmit antenna array into multiple sub-arrays, each generating its own Airy beam with a different curved trajectory, and then derive the phase offsets needed so these independently generated beams add constructively at the intended receiver. Because the signal now travels along several spatially separated paths, an obstacle-edge error that destroys one trajectory still leaves the others intact, reducing the achievable-rate loss under a 40-millimeter geometry error from 97.2% (single beam) to 33.3% (multiple beams).","feed_headline":"Multiple curved beams outperform single beam for obstacle blockage robustness","feed_subtitle":"Splitting an antenna array into sub-arrays that each generate a different Airy beam trajectory cuts geometry-error rate loss from 97% to 33%","key_machinery":"The mechanism has three load-bearing parts. First, the sensitivity bound (Equation 55): the relative change in received energy under a geometry error is controlled by a factor χ that is the ratio of the incident field energy inside the error strip to the nominal received energy. For a single Airy beam, the error strip can overlap the dominant main lobe, making χ large and the denominator small. Second, the phase-alignment condition (Theorem 1): the maximum coherent-combining power is achieved when all beam responses share the same phase, with offsets δ_m = φ_0 − φ_m. Third, the multi-Airy robustness bound (Equation 58): by the triangle inequality, the multi-beam sensitivity factor χ_M is a *","core_discovery":"The central discovery is a phase-alignment rule (Theorem 1, Equation 29) showing that M independently generated Airy beams can be coherently combined at a target point by assigning each a constant phase offset equal to the negative of its individual response phase. This rule is general—it does not depend on how the beams are generated or where their source arrays are located—and it enables a constructive beamforming procedure: partition the array, assign each sub-array a tailored Airy trajectory, compute the phase offsets, and combine. The robustness improvement is then bounded analytically (Equations 55 and 58) by showing that a geometry-error strip at the obstacle plane intersects only a *","pith_inferences":[],"forward_implications":["The coordinated multi-Airy generation method extends directly to distributed antenna systems, since the phase-alignment rule requires no assumption that the sub-arrays share a common aperture or orientation.","The trajectory-diversity principle could generalize to other self-accelerating or structured wavefronts beyond Airy beams, wherever a single dominant path creates vulnerability to obstacle-position uncertainty.","Beam-training protocols for multi-Airy systems need development: the paper assumes trajectory parameters are found by a separate search stage, but jointly optimizing trajectory selection and phase alignment remains open.","The UPA validation (6.41 dB improvement) suggests the benefit scales with aperture dimensionality, making the approach increasingly attractive for future large planar arrays."],"fun_headline_variants":["Multi-Airy beamforming phase rule enables coherent curved-beam combining","Array partitioning into multi-Airy sub-arrays cuts geometry-error rate loss","Multi-Airy phase alignment rule outperforms single-Airy under blockage","Coordinated multi-Airy beams overcome single-Airy geometry sensitivity"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The entire analysis uses a zero-thickness, single-sided opaque screen as the blockage model, with free-space diffraction on either side. Real obstacles have finite thickness, complex shapes, multiple scattering surfaces, and material-dependent penetration—none of which are captured.","fun_headline_variants_meta":{"raw":{"variants":["Multi-Airy beamforming phase rule enables coherent curved-beam combining","Array partitioning into multi-Airy sub-arrays cuts geometry-error rate loss","Multi-Airy phase alignment rule outperforms single-Airy under blockage","Coordinated multi-Airy beams overcome single-Airy geometry sensitivity"]},"model":"glm-5.2","effort":"low","cost_usd":0.0,"raw_usage":{"total_tokens":721,"prompt_tokens":658,"completion_tokens":63,"prompt_tokens_details":null},"tokens_in":658,"tokens_out":63,"duration_ms":24698,"temperature":1.0,"reasoning_tokens":null,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-09T15:12:33.816560+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"The multi-Airy robustness advantage would collapse if the obstacle simultaneously blocks all sub-array trajectories, if the trajectories overlap at the obstacle plane (so a single error strip intersects multiple main lobes), or if sub-array partitioning reduces the nominal received energy J_M excessively. The paper states these conditions explicitly (Section V-B).","supporting_citations":[],"review_version":1}