{"id":"4fbf1835-6935-4693-a7fb-857b1646657e","arxiv_id":"2505.02251","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A liquid-metal Yagi-Uda inspired reconfigurable antenna array with per-element pattern control achieves 2.5 dB beamforming gain over a fixed-pattern array in full-wave simulation, and an EM-domain channel model is formulated for it.","lead":"This paper designs an antenna array whose elements can reshape their radiation patterns using liquid metal, and it writes a wireless channel model for that array. Full-wave simulations show 2.5 dB more gain in one beamforming direction than a conventional fixed-pattern array, which may matter for future 6G beamforming systems.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (7)/(14) never defines the radiation-pattern vector g as complex field patterns with a phase reference; if magnitude-only gains were used, the channel model and its claimed full-wave agreement are invalid.","rationale":"I agree with the reader that the weakest assumption is the nature and phase reference of the pattern functions in Eq. (7). This is more load-bearing than the baseline-choice issue because it determines whether the derived channel model is mathematically valid at all. The paper's only evidence is a single HFSS comparison with no error metric or independent measurement, so the ambiguity cannot be resolved from the text. The proposed check would settle it. The conditional verdict is appropriate: no rejection is warranted without evidence that the model is wrong, but the revision must clarify and, ideally, quantify the phase convention and validation error.","tokens_in":14529,"tokens_out":12445,"duration_ms":163986,"concrete_test":"Reproduce the 135° beampattern calculation from Eq. (14) twice: (i) using the complex (magnitude-and-phase) far-field patterns exported from the HFSS element simulations with a documented phase reference, and (ii) using only the magnitude/realized-gain patterns. Compare each to the full-wave 12-element array pattern of Fig. 6, reporting the maximum and mean absolute error in dB over the full angular sweep. If case (ii) fails to match while case (i) matches, the model's validity hinges entirely on the unstated complex-pattern convention, which the revision must state explicitly; if both match, the phase-reference concern is resolved and only a clarifying sentence is needed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section III-B defines GT,i(θ) and GR,j(ϕ) as 'radiation patterns' (Eq. (4)-(5)) but never states that they are complex field patterns with a well-defined phase reference; the preceding Eq. (2) uses 'antenna gains' as real scalars. In Eq. (7), the Hadamard product gT(θ)⊙aT(θ) and the beampattern formula Eq. (14) are only physically correct if each g entry is the full complex far-field response, including direction-dependent phase relative to the element phase center, and if the same phase convention holds across all reconfigurable states. A Yagi-Uda-like element whose directors/reflector are moved will have a phase center and pattern phase that change with state; discarding this phase (e.g., using realized-gain magnitude) would make Eq. (7) and Eq. (14) invalid for coherent beamforming and for the multipath channel. The paper does not report which quantity was exported from HFSS, does not define the phase reference, and does not give a quantitative error between the model and the full-wave array pattern. The claimed 'good agreement' therefore depends on an unstated, load-bearing assumption; if that assumption is false, both the EM-domain channel model and the 2.5 dB gain validation are unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a design for electromagnetically reconfigurable antennas (ERAs) based on liquid-metal Yagi-Uda-like elements, arranges them into a 12-element uniform linear array, and derives an EM-domain channel model that extends the conventional Saleh-Valenzuela model by allowing each element to select from a dictionary of radiation patterns. The model is expressed in Eqs. (7)-(14) using Hadamard and Kronecker products, with binary selection matrices B and D. The authors validate the concept with full-wave HFSS simulations of one beamforming direction (135°) and report a 2.5 dB gain improvement over a conventional fixed-pattern array, as well as qualitative 'good agreement' between the derived beampattern and full-wave results.","tokens_in":14770,"tokens_out":3150,"duration_ms":43728,"significance":"If the central claims hold, the ERA concept introduces a practically relevant degree of freedom—per-element pattern reconfiguration—that complements the spatial reconfigurability of fluid antenna systems, and the EM-domain channel model in Eq. (12) provides a compact analytical framework for such arrays. The use of full-wave simulation as evidence is appropriate, and the model derivation is a clean reformulation of prior EM-domain channel work (Ref. [21]) adapted to element-wise pattern selection. However, the current evidence base is thin: only one beamforming angle is simulated, no operating frequency or array geometry is reported, no quantitative error metric is given for the model-versus-simulation comparison, and the complex-field phase-reference assumption underlying Eqs. (7)-(14) is never stated. The significance of the 2.5 dB gain claim is therefore not yet established beyond a single anecdotal case.","major_comments":[{"comment":"The channel model and beampattern formula are only physically correct if the quantities G_T,i(θ) and G_R,j(ϕ) are complex far-field field patterns with a well-defined phase reference (e.g., relative to each element's phase center) and if that phase convention is consistent across reconfigurable states. The paper never states this. Eq. (2) uses real scalar 'antenna gains,' and the text in Section III-B uses 'radiation patterns' without specifying complex magnitude and phase. If magnitude-only realized gains were exported from HFSS and substituted into Eqs. (7), (10), or (14), the Hadamard product with the array response vector and the coherent beamforming calculation would be invalid. The authors must specify the exact complex field quantity exported from HFSS, define the phase reference for each element state, and confirm that the same convention is used for all states. This is load-bearing for the claimed model-to-simulation agreement.","section":"Section III-B, Eqs. (4)-(7) and (14)"},{"comment":"The validation of the derived model against full-wave simulation is not quantitative. The text says 'good agreement is observed' but gives no error metric (e.g., normalized RMSE, peak beamforming error, or side-lobe level error) and no numerical comparison beyond the single quoted gain value. The figures referenced (Fig. 6) are not shown in the text, so the reader cannot assess the agreement. Please provide a quantitative comparison, ideally with multiple beamforming directions, and report the operating frequency, element spacing, and array geometry so that the comparison is reproducible. Without this, the central validation claim is unsupported.","section":"Section IV, beampattern comparison"},{"comment":"The 2.5 dB improvement over the conventional array is based on a single beamforming angle (135°) and a single benchmark state (state 2). It is unclear whether the improvement arises from genuine pattern-reconfiguration benefits or from differences in element realized gain, impedance matching, or efficiency between the states. The paper does not report the realized gain, total efficiency, or reflection coefficients of the ERA element in each state, nor does it specify whether the comparison fixes the same excitation amplitudes (it only states feeding phases are the same). Please report these quantities and clarify whether the benchmark array uses the same element spacing and the same total radiated power, so that the 2.5 dB claim is a fair comparison.","section":"Section IV, 2.5 dB gain claim"}],"minor_comments":[{"comment":"The captions contain the typo 'EAR' instead of 'ERA' (Figs. 3 and 4).","section":"Fig. 3 and Fig. 4 captions"},{"comment":"The statement 'a one-dimensional (1D) array can be formed with multiple configurations' is ambiguous; please clarify whether the reconfigurable element itself can produce different 1D array geometries or whether the 1D array is fixed while only the element patterns change.","section":"Section II, array description"},{"comment":"Reference [16] is cited as having 'adopted and extended' the same model, and Ref. [11] describes a closely related architecture. To clarify the novel contribution, please state explicitly what this paper adds beyond [16] and [11] in terms of hardware design, validation, and channel-model formulation.","section":"Section I and reference [16]"},{"comment":"The operating frequency, the dimensions of the ERA element, the inter-element spacing dI, and the substrate parameters are not reported anywhere in the text. Including them is essential for reproducibility and for assessing whether the half-wavelength element-size constraint mentioned in the introduction is met.","section":"Section IV, simulation parameters"},{"comment":"The notation ||b||_2 = 1 for a binary vector with exactly one nonzero entry is unconventional; using ||b||_2^2 = 1 or the explicit one-hot constraint would be clearer.","section":"Section III-B, Eq. (6)"}],"recommendation":"major_revision","confidential_remarks":"The paper reads like an extended abstract or a preliminary report. The core idea is interesting, but the experimental/applied validation is minimal: one beamforming angle, no quantitative model comparison, and no hardware. The missing complex-phase specification in the channel model is a technical point that the authors can likely fix by clarifying their HFSS export, but it is load-bearing. I would advise the editor that the manuscript needs substantial additional simulation or measurement data before it can be considered for publication, even though the direction is promising."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The two things you should know: this is a concept paper, not a hardware demonstration, and its main new contribution is the combination of a liquid-metal Yagi-Uda-inspired reconfigurable element with a compact selection-matrix channel model. The 2.5 dB gain over a fixed-pattern array is a real simulated result, though only at one beamforming angle with no uncertainty quantification. The channel model in Eq. (7)-(13) is a clean specialization of the EM-domain channel of [21], and the block-diagonal selection matrices B and D are a tidy way to bookkeep per-element pattern choices. That is genuinely useful for system-level optimization, and the paper deserves credit for making the array element physically plausible with microfluidics and a monopole feed.\n\nNow the soft spots, in proportion. The stress-test concern is real and load-bearing: Section III never states that the pattern functions GT,i(θ) and GR,j(ϕ) are complex field responses with a defined phase reference. Equation (2) calls them 'antenna gains' (real scalars), and Eq. (7) and (14) only make sense if each entry of g is the full complex far-field response including direction-dependent phase relative to a common phase center. A Yagi-Uda-like element with moving directors/reflectors will have a phase center that shifts with state. If HFSS exports realized-gain magnitude only, then the Hadamard-product factorization and the claimed agreement are invalid. The paper does not say which quantity was exported, does not define the phase reference, and gives no quantitative error metric between the model and full-wave simulation. 'Good agreement is observed' is the only evidence offered.\n\nOther weaknesses are minor by comparison: only one beamforming angle (135°) is shown, no operating frequency or element spacing is given, and mutual coupling is ignored when moving from the single element to the 12-element array. Those are the usual first-round revision requests, not fatal flaws. The gain claim itself comes directly from HFSS, so there is no circularity in that specific number.\n\nWho is this for? Researchers working on reconfigurable antennas for 6G, especially those wanting a tractable channel model for element-reconfigurable arrays. It is not a definitive experimental validation, but it is a legitimate design plus modeling contribution. A serious referee should see it, with requests to clarify the complex-pattern definition, add a quantitative agreement metric, and ideally extend the study to a second beamforming angle. My verdict: conditional acceptance with major revisions, not rejection.","headline":"A plausible liquid-metal element-reconfigurable array with a clean channel-model bookkeeping trick, but the claimed full-wave 'good agreement' leans on an unstated assumption about complex pattern phases.","tokens_in":15323,"tokens_out":1230,"would_cite":false,"duration_ms":19422,"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":"An electromagnetically reconfigurable antenna array is claimed to achieve 13.5 dBi at 135 degrees, 2.5 dB above a conventional array, with an EM-domain channel model matching full-wave simulation.","keywords":["electromagnetically reconfigurable antennas","fluid antenna systems","liquid metal antennas","radiation pattern reconfigurability","MIMO channel modeling","beamforming","full-wave simulation","Yagi-Uda antenna"],"falsifier":"Simulate or measure the full 12-element array with several different per-element pattern states and compare the realized beampattern with the one predicted by Eq. (14) using isolated element patterns; if the disagreement at the main beam is comparable to the claimed 2.5 dB gain margin, the model's no-coupling assumption fails.","tokens_in":1803,"feed_emoji":"📡","tokens_out":1946,"duration_ms":100852,"temperature":0.7,"pith_summary":"The paper tries to establish that antenna arrays gain a useful new degree of freedom when each element can change its own radiation pattern, not just its position or excitation. It proposes a practical element built around liquid-metal directors and a reflector, arranged like a miniature Yagi-Uda antenna, and shows in full-wave simulation that a 12-element version of this array outperforms a conventional fixed-pattern array at beamforming. The load-bearing quantitative claim is a 13.5 dBi realized gain at 135 degrees, 2.5 dB higher than the conventional array, along with 5.5 dB of sidelobe suppression. The paper also derives an EM-domain channel model in which selecting each element's pattern enters through binary selection matrices, and it reports that beampatterns computed from that model agree with the full-wave simulation. A sympathetic reader would care because this points toward hardware that can reshape the wireless channel itself, with a tractable model for system design.","feed_headline":"Reconfigurable antenna array gains 2.5 dB over fixed design","feed_subtitle":"Liquid-metal directors reshape each element's pattern; a new EM-domain channel model reproduces the simulated beam.","key_machinery":"The carrying object is the ERA element, a Yagi-Uda-inspired structure: a planar monopole excites six parallel liquid-metal tubes acting as directors plus a larger rear liquid-metal element acting as reflector, and fluid length and position reconfigure the element's radiation pattern continuously. Around this hardware, the model's central mathematical object is the pattern dictionary vector $\\bar{\\mathbf{g}}(\\varphi)$ together with the block-diagonal selection matrices $\\mathbf{B}$ and $\\mathbf{D}$, which convert element-wise pattern choice into the channel expression $\\mathbf{H}_{\\mathrm{ER}} = \\gamma \\mathbf{D} \\mathbf{H}_{\\mathrm{EM}} \\mathbf{B}^{T}$. Here $\\mathbf{H}_{\\mathrm{EM}}$ is the EM-domain channel: a sum over line-of-sight and scattered paths of Kronecker products $(\\mathbf{a}_R \\otimes \\bar{\\mathbf{g}})(\\mathbf{a}_T \\otimes \\bar{\\mathbf{g}})^{H}$, where the Kronecker product pairs each array response with the reconfigurable pattern dictionary. This object does the work of making pattern reconfigurability appear inside the channel matrix rather than outside it as a scalar gain, and Eq. (14) uses the same object to synthesize beampatterns that can be checked against full-wave simulation.","core_discovery":"The paper's central claim is that electromagnetic reconfigurability of individual array elements is a real and useful extension of fluid antenna systems. An ERA element can be implemented with a planar monopole exciter, six liquid-metal directors, and a rear liquid-metal reflector, all continuously adjustable, so the element's complex radiation pattern can be steered. When twelve such elements form a uniform linear array and are driven with the same phase differences as a conventional array, the full-wave simulated realized gain at the 135 degree beamforming direction is 13.5 dBi, which is 2.5 dB higher than the fixed-pattern benchmark, while the main sidelobe drops from 6.7 dBi to 1.2 dBi. The paper further claims that the channel formed by such an array is captured by $\\mathbf{H}_{\\mathrm{ER}} = \\gamma \\mathbf{D} \\mathbf{H}_{\\mathrm{EM}} \\mathbf{B}^{T}$, where $\\mathbf{B}$ and $\\mathbf{D}$ are block-diagonal selection matrices encoding each element's chosen pattern and $\\mathbf{H}_{\\mathrm{EM}}$ is an EM-domain channel built from array response vectors and the pattern dictionary; calculated beampatterns from this model agree with full-wave simulation. The conclusion the authors draw is that the ERA concept and its channel model are validated by full-wave EM simulation.","pith_inferences":["If the no-coupling assumption holds in practice, array-level beampatterns can be predicted from isolated element patterns alone, which would make optimization over per-element pattern states computationally cheap.","A natural testable extension is to fabricate the liquid-metal element and measure embedded element patterns; the claimed 2.5 dB gain margin may shrink when mutual coupling and microfluidic tolerances are included.","The same EM-domain channel structure could be applied to frequency- or polarization-reconfigurable elements, since those would only change the dictionary vectors and selection matrices.","The paper names near-field beam focusing and large-angle scanning as possible applications; the channel model as written is far-field, so extending it to near-field would require replacing planar-wave array responses with spherical-wave ones."],"forward_implications":["At a 135 degree beamforming direction, the ERA array is claimed to achieve 13.5 dBi realized gain, 2.5 dB above a conventional array with the same feeding phases.","The main sidelobe at 45 degrees is suppressed from 6.7 dBi to 1.2 dBi, a 5.5 dB reduction, so the reconfigurable elements concentrate power in the desired direction.","The derived channel model with selection matrices gives a tractable way to include element-level pattern reconfigurability in MIMO signal models with a single RF chain and analog phase shifters.","Because each element's pattern set is a dictionary, the same model can be reused for different reconfigurable element designs by swapping the dictionary vectors.","The reported agreement between model and full-wave simulation means system-level studies can use the model rather than simulating every configuration."],"supporting_citations":[{"why":"Provides the director-reflector antenna concept that the ERA element is based on.","marker":"[14]"},{"why":"Supplies the compact planar microstrip-fed Yagi topology used for the exciter design.","marker":"[15]"},{"why":"Demonstrates liquid metal together with 3-D-printed microfluidics as the reconfiguration mechanism.","marker":"[9]"},{"why":"Supplies the multipath channel model that Eq. (2) extends to the reconfigurable case.","marker":"[17]"},{"why":"Provides a channel-simulator basis for the conventional-array channel expression.","marker":"[18]"},{"why":"Introduces the electromagnetic-domain channel concept that Eq. (13) adapts to element-pattern reconfigurability.","marker":"[21]"},{"why":"Gives the beampattern synthesis expression used in Eq. (14) to verify the model.","marker":"[22]"},{"why":"Provides the ERA-based beampattern synthesis formulation used for comparison with full-wave simulation.","marker":"[23]"}],"fun_headline_variants":["ERA array beats fixed design by 2.5 dB","Electromagnetic reconfigurability adds 2.5 dB gain","Fluid antenna elements yield 2.5 dB beam gain","Reconfigurable elements outdo static array by 2.5 dB","Shaping each antenna element pays off with 2.5 dB"],"cache_read_input_tokens":17408,"weakest_assumption_plain":"The model and the claimed agreement with simulation assume that each element keeps its isolated radiation pattern inside the 12-element array, so electromagnetic coupling between elements is not accounted for.","fun_headline_variants_meta":{"raw":{"variants":["ERA array beats fixed design by 2.5 dB","Electromagnetic reconfigurability adds 2.5 dB gain","Fluid antenna elements yield 2.5 dB beam gain","Reconfigurable elements outdo static array by 2.5 dB","Shaping each antenna element pays off with 2.5 dB"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000196,"raw_usage":{"total_tokens":1350,"prompt_tokens":921,"completion_tokens":429,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":537,"completion_tokens_details":{"reasoning_tokens":340}},"tokens_in":537,"tokens_out":429,"duration_ms":4620,"temperature":1.0,"reasoning_tokens":340,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:57:28.279434+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate or measure the full 12-element array with several different per-element pattern states and compare the realized beampattern with the one predicted by Eq. (14) using isolated element patterns; if the disagreement at the main beam is comparable to the claimed 2.5 dB gain margin, the model's no-coupling assumption fails.","supporting_citations":[{"cited_title":"Beam transmission of ultra short waves,","cited_arxiv_id":null,"evidence_quote":"Provides the director-reflector antenna concept that the ERA element is based on."},{"cited_title":"Compact planar microstrip-fed quasi-yagi antenna,","cited_arxiv_id":null,"evidence_quote":"Supplies the compact planar microstrip-fed Yagi topology used for the exciter design."},{"cited_title":"Wideband frequency reconfigurable patch antenna with switchable slots based on liquid metal and 3-D printed microfluidics,","cited_arxiv_id":null,"evidence_quote":"Demonstrates liquid metal together with 3-D-printed microfluidics as the reconfiguration mechanism."},{"cited_title":"Spatially sparse precoding in millimeter wave MIMO systems,","cited_arxiv_id":null,"evidence_quote":"Supplies the multipath channel model that Eq. (2) extends to the reconfigurable case."},{"cited_title":"Reconfigurable massive MIMO: Precoding design and channel estimation in the electromagnetic domain,","cited_arxiv_id":null,"evidence_quote":"Introduces the electromagnetic-domain channel concept that Eq. (13) adapts to element-pattern reconfigurability."},{"cited_title":"Mutual coupling in RIS-aided communication: Model training and experimental validation,","cited_arxiv_id":null,"evidence_quote":"Gives the beampattern synthesis expression used in Eq. (14) to verify the model."}],"review_version":1}