{"id":"d4759eba-7146-4bb8-bc07-3178ef37c95f","arxiv_id":"2506.17559","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Closed-form average SNRs and gain thresholds for three coordination levels of joint base-station and pinching-antenna transmission show full cooperation always helps while the cheaper schemes help only under favorable geometry.","lead":"This paper proposes three levels of cooperation between a cellular base station and flexible pinching antennas clipped onto waveguides, and derives closed-form formulas for the average signal quality each level achieves. The formulas give network planners concrete thresholds for when pinching antennas help and when they do not.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Uniform-phase averaging is load-bearing for SD/SCD gains; fixed deployments retain a geometry-dependent BS-PAS cross term that can reverse the Remark 1 thresholds.","rationale":"The reader's weakest assumption is exactly the load-bearing issue I find: the uniform-random-phase model is introduced in Proposition 1 and Appendix A, used to zero the cross term in Eq. (31), and baked into the Section V simulation by randomizing the PAS-UE distance. My independent check of the algebra confirms that the Table III expressions are correct under that model; the problem is not internal consistency but scope. For a fixed PAS geometry the phases are deterministic, the cross term does not vanish, and the average SNR gains of SD and SCD become geometry-dependent and can be much worse than the closed forms predict. This is especially consequential for SCD, where the paper claims (Remark 1, Example 4) that SCD always beats BS-only for alpha >= 2 in the typical scenario; a fixed phase near pi can make the coherent PAS-UE and BS-UE contributions cancel on average. The FCD result, by contrast, is a sum of independent nonnegative terms and is robust to the phase model. I therefore do not move the verdict: CONDITIONAL remains appropriate, and the paper should either state that its SD/SCD results are for a random-geometry ensemble or add a fixed-deployment analysis with explicit phase dependence.","tokens_in":20735,"tokens_out":11956,"duration_ms":138791,"concrete_test":"Run a fixed-deployment Monte Carlo with Table V parameters (N_B=64, K=4, N_G=8, L_B=200 m, L_G=100 m, alpha=2.4, beta=2): keep phi_1 = ... = phi_K = pi fixed across trials, draw only the BS Rayleigh fading vector tilde_h_B ~ CN(0,I) for 10^5 trials, and average |h_S|^2 and |h_C|^2 using the exact beamformers of Eqs. (10) and (16). Compare against the Table III predictions and against the BS-only average. If the fixed-phase SCD average drops below the BS-only average, or differs from Eq. (18) by more than 20%, the closed forms do not describe a fixed deployment and the Remark 1/Example 4 'always outperforms' statement is conditional on the random-phase ensemble.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The SD and SCD closed forms in Table III are derived by dropping the BS-PAS cross term in Eq. (31) via the assumption in Proposition 1 and Appendix A that each reference-antenna phase phi_k is i.i.d. uniform on [0, 2*pi]. That is an ensemble assumption, not a property of a fixed installation: in a deployed geometry phi_k is a deterministic function of positions through Eq. (2). If only the Rayleigh fading is averaged, a fixed deployment retains the cross term 2*sqrt(eta*P_S,B/L_B^alpha)*sqrt(eta*N_G*P_S,G)*E[||tilde_h_B||]*Re(sum_k sqrt(1/L_G,k^beta)*e^{-j*phi_k}) for SD, and the analogous term 2*sqrt(eta*P_C,B/L_B^alpha)*sqrt(eta*P_C,G*N_G*K/L_G^beta)*E[||tilde_h_B||]*cos(phi_1) for SCD. This term scales with sqrt(N_B) and can be positive or negative; with the Section V parameters and phi_1 = pi the SCD cross term nearly cancels the two positive terms, so the achievable average SNR can fall below the BS-only value even in the regime where Example 4 and Remark 1 predict SCD always wins. The simulation in Section V makes the assumption true by drawing the PAS-UE distance uniformly in [L_G - lambda/2, L_G + lambda/2], so the Monte Carlo verification is circular with respect to the fixed-deployment reading. Consequently, the SD/SCD rows of Table III and the gain thresholds in Remark 1 describe a random-geometry ensemble rather than the fixed PAS deployments implied by the title and abstract; FCD, whose average is a sum of independent positive terms, is not affected by this concern.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes three BS-PAS joint transmission schemes for downlink cellular networks—standalone deployment (SD), semi-cooperative deployment (SCD), and full-cooperative deployment (FCD)—and derives closed-form average received SNR expressions for each (Table III, Eqs. (13), (19), (26)). The derivations assume MRT-type beamforming, specific power-splitting coefficients (Table II), and a Rayleigh-faded BS-UE link with LoS PAS-UE links whose reference-antenna phases are i.i.d. uniform on [0,2π]. The paper also derives gain ratios over a BS-only baseline (Table IV) and gives conditions (Remark 1, Examples 4–5) under which SD and SCD outperform BS-only, while FCD always does. Monte Carlo simulations are presented as verification. The central analytical work is internally consistent under the stated phase-ensemble model, but the SD and SCD results rely crucially on the uniform-phase assumption, which is imposed in the simulations rather than tested against fixed deployment geometries.","tokens_in":21012,"tokens_out":6589,"duration_ms":68252,"significance":"If taken as an analysis of a random-geometry ensemble of pinching-antenna placements, the paper offers useful, parameter-free closed-form formulas and simple design thresholds for when PAS cooperation helps. The FCD result, which averages a sum of independent positive terms, is robust and does not depend on the problematic phase-averaging; the unconditional FCD gain V_FCD > 1 is a clean and useful insight. The derivations are self-contained, with no fitted parameters; I re-derived Eqs. (12), (18), (25) and the Table III/IV entries and found no algebraic errors under the stated model. The main limitation is that the SD/SCD closed forms and the Remark 1 thresholds describe an ensemble of uniformly random PAS phases, not a fixed installation, and the simulation does not independently validate the fixed-deployment reading promised by the title and abstract.","major_comments":[{"comment":"The SD and SCD average-SNR formulas (Eqs. (12) and (18)) are derived by dropping the BS–PAS cross term in Eq. (31) via E{cos Ω}=0, which requires φ_k to be i.i.d. uniform on [0,2π]. In a fixed deployment, φ_k is deterministic through Eq. (2), so the cross term does not vanish when the expectation is taken only over the Rayleigh fading. For SD the retained term is 2√(η P_S,B/L_B^α) √(η N_G P_S,G) E[‖h̃_B‖] Re(Σ_k √(1/L_G,k^β) e^{-jφ_k}), and for SCD the analogous term is proportional to cos(φ_1). This term scales with √N_B and can be negative; with the Section V parameters and φ_1=π it can nearly cancel the two positive SCD terms, so the achievable average SNR can fall below BS-only even in a regime where Remark 1 and Example 4 predict SCD always wins. The thresholds in Remark 1 and Table IV are therefore ensemble averages over random geometry, not guarantees for a specific PAS installation. Please either (i) explicitly scope the SD/SCD claims to the random-phase ensemble throughout the title, abstract, and conclusions and add a true fixed-geometry simulation, or (ii) derive the fixed-geometry expressions that retain the geometry-dependent cross term and revisit the Remark 1 conditions.","section":"Section III-B / Appendix A / Propositions 1 and 3"},{"comment":"The Monte Carlo verification is circular with respect to the fixed-deployment interpretation. The simulation generates the PAS–UE distance uniformly in [L_G−λ/2, L_G+λ/2] precisely to make each reference-antenna phase uniform on [0,2π], which is the same assumption used in Appendix A and Proposition 1 to eliminate the BS–PAS cross term. Consequently Fig. 3 and the other simulation figures validate that the closed forms match the ensemble model, but they do not test whether the formulas describe a particular installed geometry. Please add a simulation with fixed, deterministic PAS positions (e.g., several representative geometries from Example 1-type coordinates) and compare against the fixed-geometry expressions, or clearly label the current simulation as an ensemble-consistency check.","section":"Section V (simulation setup)"}],"minor_comments":[{"comment":"The third term in Eq. (31) is written with P_S,B, but from the subsequent derivation it should be P_S,G; also the cross term in Eq. (31) appears to be missing the factor 2 from |a+b|^2. The final result is unaffected, but the typo should be corrected.","section":"Eq. (31)"},{"comment":"The SCD beamforming vector in Eq. (16) is typeset in a garbled way; the power-normalization factor and the vector entries are not clearly separated. Please rewrite the expression so that the normalization by √(Σ_k η N_G/L_G,k^β) is explicit.","section":"Eq. (16)"},{"comment":"The proof that Ω is uniform on (0,2π) is not fully rigorous: the conditional-PDF notation and the normalization in step 3 are unclear, and the argument that ∫ F(ΔΦ)dΦ = ∫ F(ΔΦ)dω_1 = 1 is stated without defining F as a joint density. The result is true for independent uniform phases, but the proof would benefit from a cleaner statement using the rotation-invariance of a single uniform phasor.","section":"Appendix A"},{"comment":"There are numerous typographical and OCR-style errors: 'vanilla example', 'givn', 'W aveguide', 'greaterorequalslant', 'sufficient', and the duplicated phrase 'practical insights practical insights'. A careful proofreading pass is needed.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The central derivations are correct under the stated ensemble model, and the FCD analysis is robust. The load-bearing issue is the uniform-phase assumption for SD/SCD and the circular simulation. In my view this is fixable within the manuscript's scope by re-scoping the claims to a random-geometry ensemble or by adding fixed-geometry cross-term analysis; it is not an unfixable error. I would therefore not reject but would require a major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThis paper proposes a three-level cooperation taxonomy for joint BS-pinching-antenna transmission (standalone, semi-cooperative, full-cooperative), with MRT beamforming and closed-form average SNR expressions for each. The math is solid: I rechecked the derivations in Propositions 1 and 3 and the gain comparisons in Table IV; they are correct under the stated model. The FCD result is exact for any phase realization because MRT co-phases everything. The taxonomy and the closed forms are genuinely new relative to the PAS literature, and the paper honestly identifies regimes where SD and SCD lose to BS-only.\n\nThe soft spot is the uniform-random-phase assumption that SD and SCD closed forms rely on. Proposition 1 states it, and Appendix A uses it to drop the BS-PAS cross term. But in a fixed deployment, the phases are deterministic functions of the geometry; only the Rayleigh fading is random, and the cross term 2·sqrt(...)·||h_B||·cos(Ω) remains. With the paper's own Table V parameters and φ_1 = π, the SCD cross term nearly cancels the positive terms, so the average SNR can fall below BS-only even where Remark 1 predicts SCD always wins. The simulation manufactures the assumption by drawing the PAS-UE distance uniformly, so it is a consistency check on the ensemble model, not validation for a fixed deployment. FCD is not affected; only the SD/SCD formulas and the Remark 1 thresholds lose their direct practical meaning.\n\nMinor issues: Example 6 claims V_FCD ≈ 1 + 1224/N_B, but with Table V parameters the coefficient is about 1066. Eqs. (22)-(23) present an average-of-ratios as if it were a ratio of averages, an approximation not flagged as such. Neither affects the main SNR expressions.\n\nThis deserves a serious referee. A careful revision should add a fixed-deployment analysis or at least a bound on the cross term, and state in the abstract that SD/SCD results are ensemble averages. As is, I would not cite it for quantitative thresholds, but the taxonomy and the FCD result are worth knowing.","headline":"Solid math and a genuinely useful taxonomy, but the SD/SCD closed forms average over random phases that a fixed deployment never has; FCD is robust.","tokens_in":21693,"tokens_out":8129,"would_cite":false,"duration_ms":72704,"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":"The paper derives closed-form average SNR for three BS-pinching-antenna transmission schemes and shows only full cooperation always beats the BS-only baseline.","keywords":["pinching antennas","joint transmission","beamforming","average received SNR","distributed antenna systems","cellular networks","path loss"],"falsifier":"Fix one deployment geometry with non-random reference-antenna phases and average only over the Rayleigh fading of the BS channel; if the simulated SD or SCD SNR disagrees with Table III, the uniform-phase assumption is load-bearing and the formulas describe a random-phase ensemble rather than that fixed installation.","tokens_in":20404,"feed_emoji":"📶","tokens_out":11922,"duration_ms":105982,"temperature":0.7,"pith_summary":"The paper proposes three ways to coordinate a cellular base station with waveguide-mounted pinching antennas — small radiating elements clipped onto a dielectric waveguide — and derives closed-form expressions for the average received SNR that each coordination level achieves. The central result is a set of simple ratio rules: full cooperative deployment always beats the base-station-only baseline, while standalone and semi-cooperative deployments improve on it only when the BS-to-user path loss is sufficiently worse than the pinching-antenna-to-user path loss. These formulas matter because pinching antennas are cheap and flexible, so an operator deciding whether to deploy them, and how much coordination to invest in, gets concrete numerical thresholds rather than a qualitative promise. The paper also identifies how the gains scale with the numbers of BS antennas, waveguides, and pinching antennas per waveguide.","feed_headline":"Pinching antennas always help when BS and waveguides fully cooperate","feed_subtitle":"Closed-form SNR formulas show when standalone and semi-cooperative setups beat the base station alone.","key_machinery":"The central object is the aggregate MISO channel from the $N_B+K$ transmit ports, $$h=\\left[\\sqrt{\\frac{\\eta}{L_B^\\$\\alpha$}}\\tilde{h}_B,\\ \\sqrt{\\frac{\\eta N_G}{L_{G,1}^{\\$\\beta$}}}$e^{{-j\\varphi_1}}$,\\ \\ldots,\\ \\sqrt{\\frac{\\eta N_G}{L_{G,K}^{\\$\\beta$}}}$e^{{-j\\varphi_K}}$\\right],$$ which treats each waveguide as one RF port after its $N_G$ pinching antennas are placed to satisfy the phase-coherence condition (4). The argument then runs on maximum-ratio transmission and three power-allocation rules: equal per-port power in SD, waveguide power allocation proportional to channel gains in SCD, and full dynamic allocation in FCD. The uniform-phase assumption is what zeroes the BS-PAS cross term in the SD and SCD averages; in FCD, beamforming phase-aligns the whole channel, so the same averaging assumption is not needed.","core_discovery":"On its own terms, the paper's central claim is that, for a user whose line-of-sight to the BS is blocked, the average received SNR under maximum-ratio transmission is given in closed form for three coordination architectures. With $K$ waveguides each carrying $N_G$ pinching antennas at common distance $L_G$, and $N_B$ BS antennas at distance $L_B$, the average SNRs are $$\\gamma_{\\mathrm{SD}} = \\left(\\frac{\\eta $N_B^{2}$}{L_B^\\$\\alpha$(N_B+K)}+\\frac{\\eta N_G K}{L_G^\\$\\beta$(N_B+K)}\\right)\\tilde{\\gamma},$$ $$\\gamma_{\\mathrm{SCD}} = \\left(\\frac{\\eta $N_B^{2}$}{L_B^\\$\\alpha$(N_B+K)}+\\frac{\\eta N_G $K^{2}$}{L_G^\\$\\beta$(N_B+K)}\\right)\\tilde{\\gamma},$$ $$\\gamma_{\\mathrm{FCD}} = \\left(\\frac{\\eta N_B}{L_B^\\$\\alpha$}+\\frac{\\eta N_G K}{L_G^\\$\\beta$}\\right)\\tilde{\\gamma}.$$ The FCD gain over the BS-only baseline is $V_{\\mathrm{FCD}}=1+(N_GK/N_B)(L_B^\\alpha/L_G^\\beta)>1$; the SD and SCD gains exceed 1 only when $L_B^\\alpha/L_G^\\beta>N_B/N_G$ and $L_B^\\alpha/L_G^\\beta>N_B/(N_GK)$, respectively. These expressions are derived under the assumption that each waveguide's reference phase $\\varphi_k$ is uniformly distributed over $[0,2\\pi]$, and the paper's Monte Carlo simulations randomize the reference-antenna distance within half a wavelength to produce exactly that condition.","pith_inferences":["Because the SD and SCD formulas average over random waveguide phases, a network operator evaluating a specific mounting position should either use FCD or account for the BS-PAS cross term, rather than reading Table III as a per-site prediction.","The computed thresholds suggest a cheap pre-deployment test: measure the path-loss ratio $L_B^\\alpha/L_G^\\beta$; if it sits below $N_B/N_G$ or $N_B/(N_GK)$, adding waveguides without full cooperation should lower average SNR.","In a multiuser scenario the single-user phase-coherence condition cannot hold for all users at once, so the same gain formulas would likely become scheduling- and overhead-dependent; this extension is not studied in the paper.","The same closed-form style could be carried over to uplink or NOMA-assisted pinching-antenna systems, where the thresholds would depend on per-user distances and power budgets; the paper does not claim this."],"forward_implications":["In the path-loss-dominated regime $L_B^\\alpha\\gg L_G^\\beta$, the SCD gain is $K$ times the SD gain, and the FCD gain is an additional factor $1+N_B/K$ over SCD.","When the BS-UE path loss is not sufficiently larger than the PAS-UE path loss, standalone or semi-cooperative deployment can be worse than doing nothing, because power is diverted from the BS without enough pinching-antenna beamforming gain to compensate.","The FCD gain grows linearly in the total number of pinching antennas $N_G K$ and in the path-loss ratio $L_B^\\alpha/L_G^\\beta$, so full cooperation is the only architecture whose benefit over BS-only operation is unconditional.","As the BS antenna count grows, all three joint schemes asymptotically match the BS-only scheme; with the paper's typical parameters the FCD advantage falls to 3 dB only near $N_B\\approx 1224$.","As the number of waveguides grows, the SD gain saturates while the SCD and FCD gains grow linearly with $K$, meaning inter-waveguide cooperation is what turns additional waveguides into array gain."],"supporting_citations":[{"why":"Supplies the pinching-antenna concept and the LoS channel model for a waveguide-mounted pinching antenna used throughout.","marker":"[7]"},{"why":"Justifies neglecting transmission loss inside the waveguide, so the PAS-UE channel is path loss times an array factor.","marker":"[10]"},{"why":"Provides the clustered PAS placement and the phase-coherence condition that lets each waveguide act as a single RF port.","marker":"[12]"},{"why":"Grounds using distinct path-loss exponents for the NLoS BS-UE link and the LoS PAS-UE link.","marker":"[29]"},{"why":"Supports maximum-ratio transmission as the BS beamforming rule and the phase-noise precompensation assumption behind PAS phase alignment.","marker":"[31]"},{"why":"Supports the beam-training and phase-alignment mechanism used to tune reference antenna positions.","marker":"[33]"}],"fun_headline_variants":["Full BS-pinching cooperation always improves SNR; partial not always","Only full cooperation with pinching antennas beats BS alone in SNR","FCD always wins; SD and SCD gains depend on path-loss ratios","Closed-form SNR: full BS-PAS cooperation beats BS always, partial conditionally"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The SD and SCD closed forms require the phase of each waveguide's reference pinching antenna to be uniformly random over all angles from 0 to $2\\pi$; in a fixed installation that phase is fixed by geometry, so the averaged formulas can miss a geometry-dependent interference term between the base station and the pinching antennas.","fun_headline_variants_meta":{"raw":{"variants":["Full BS-pinching cooperation always improves SNR; partial not always","Only full cooperation with pinching antennas beats BS alone in SNR","FCD always wins; SD and SCD gains depend on path-loss ratios","Closed-form SNR: full BS-PAS cooperation beats BS always, partial conditionally"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000713,"raw_usage":{"total_tokens":3294,"prompt_tokens":1120,"completion_tokens":2174,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":736,"completion_tokens_details":{"reasoning_tokens":2095}},"tokens_in":736,"tokens_out":2174,"duration_ms":17145,"temperature":1.0,"reasoning_tokens":2095,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:11:10.600568+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fix one deployment geometry with non-random reference-antenna phases and average only over the Rayleigh fading of the BS channel; if the simulated SD or SCD SNR disagrees with Table III, the uniform-phase assumption is load-bearing and the formulas describe a random-phase ensemble rather than that fixed installation.","supporting_citations":[{"cited_title":"Pinching antenna: Using a dielectric waveguide as an antenna,","cited_arxiv_id":null,"evidence_quote":"Supplies the pinching-antenna concept and the LoS channel model for a waveguide-mounted pinching antenna used throughout."},{"cited_title":"Rate maximizat ion for downlink pinching-antenna systems,","cited_arxiv_id":null,"evidence_quote":"Provides the clustered PAS placement and the phase-coherence condition that lets each waveguide act as a single RF port."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Grounds using distinct path-loss exponents for the NLoS BS-UE link and the LoS PAS-UE link."},{"cited_title":"Phase noise degradation in massive MIMO downlink with zero-forcing and maximum ratio transmission precoding,","cited_arxiv_id":null,"evidence_quote":"Supports maximum-ratio transmission as the BS beamforming rule and the phase-noise precompensation assumption behind PAS phase alignment."}],"review_version":2}