{"id":"33003313-304f-4d8f-b97b-a06beab978e7","arxiv_id":"2508.07002","paper_version":6,"verdict":"REJECT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":3,"one_line_summary":"Jointly optimizing transmit beamforming and pinching-antenna positions in a backscatter-assisted symbiotic radio system is claimed to raise achievable sum rate by 17-35% over fixed-antenna baselines.","lead":"Combining pinching antennas, which slide along dielectric waveguides, with symbiotic radio, where a backscatter device piggybacks on the primary signal, this paper proposes two algorithms to jointly choose antenna positions and transmit beamforming. Generalist readers might care because the scheme claims large sum-rate gains over fixed-antenna baselines for future low-power IoT networks.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (10) reverses the direction of the [37] bound: the correct inequality is Pe ≥ 1 − sqrt(D/2), so constraint (11c) cannot certify Pe ≤ 1 − ε; the advertised detection guarantee is unsupported.","rationale":"The reader's REJECT is well-founded. The load-bearing step is the conversion of a detection error probability requirement into the tractable KL constraint used in both algorithms. Eq. (10) misstates the direction of the known bound from [37]; with the correct inequality, D ≥ 2ε² does not guarantee Pe ≤ 1 − ε. Since (11c) is the only place the IoT detection requirement enters the optimization, every reported sum-rate result is computed under a constraint that is not the one claimed. The SCA convergence proof also has a questionable monotonicity step: solving a lower-bound surrogate (P2.2) does not by itself ensure the true objective increases, so inequality (33)(a) is not justified. I did not select this as the primary concern because even a non-monotone SCA can be numerically convergent, whereas the direction error in (10) directly attacks the central advertised guarantee. Missing error bars and incomplete hyperparameter details are reporting issues, not fatal to the technical claim. The algorithms themselves may be salvageable after correcting (10)-(11c), but the paper as written does not establish its central claim, so no verdict adjustment is needed.","tokens_in":18886,"tokens_out":8325,"duration_ms":79260,"concrete_test":"Re-derive Eq. (10) from the cited reference [37]: locate the binary hypothesis-testing theorem and check whether it states P_FA + P_MD ≥ 1 − sqrt(1/2 D(P0||P1)) (a lower bound) rather than an upper bound. Then, using the paper's own simulation settings (K=N=2, M=3, L=60, ε=0.95, Pmax=30 dBm), run SCA-PSO and, for 100 random channel realizations, evaluate the exact likelihood-ratio error probability ξ* = Pr(B1|H0)+Pr(B0|H1) from Eq. (9) at each optimized (W,X). If ξ* exceeds 1−ε = 0.05 in a non-negligible fraction of realizations, (11c) fails to enforce the advertised detection error probability and the central claim is invalidated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In Section II-B, Eq. (10) states 'a tractable lower bound on Pe is obtained according to [37], expressed as Pe ≤ 1 − sqrt(1/2 D(P0||P1))'. This is internally inconsistent (a lower bound cannot be ≤), and the cited result in [37] is the standard converse bound for binary hypothesis testing: P_FA + P_MD ≥ 1 − sqrt(1/2 D(P0||P1)). Since Pe in Eq. (8) is defined as Pr(B1|H0)+Pr(B0|H1), i.e., P_FA+P_MD, the correct relation is Pe ≥ 1 − sqrt(D/2). Therefore the reformulated constraint (11c), D(P0||P1) ≥ 2ε², yields only Pe ≥ 1 − sqrt(D/2), a quantity ≤ 1 − ε (or negative for D > 2). It does not imply the advertised upper bound Pe ≤ 1 − ε. Because (11c) is the only mathematical encoding of the IoT receiver's detection error probability requirement, all optimized solutions from both LGD and SCA-PSO are not certified to meet the stated requirement. The abstract and conclusion's claim that the design 'satisfies the IoT receiver's detection error probability constraint' therefore rests on an invalid inequality. The rate gains in Figs. 5–9 are obtained under a constraint whose advertised interpretation is wrong, not under the strict detection constraint stated in the paper. If the authors intended Pe to be the average error probability (Pe = ξ/2), the lower bound acquires a factor 1/2 but remains a lower bound; the direction error and its consequence for (11c) persist.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper proposes a downlink symbiotic radio (SR) system in which a pinching-antenna system (PASS) base station serves multiple primary receivers and a backscatter-based IoT receiver. The authors formulate a joint transmit beamforming and pinching-antenna position optimization problem (P1) to maximize the primary sum rate subject to a detection error probability constraint at the IoT receiver, a minimum PA spacing constraint, and waveguide length limits. Two solution algorithms are proposed: a learning-aided gradient descent (LGD) method and a two-stage SCA-PSO method. Numerical results report substantial sum-rate gains of the PASS-SR design over fixed-antenna, conventional MIMO, and massive MIMO baselines.","tokens_in":19305,"tokens_out":4807,"duration_ms":52467,"significance":"If the central detection-constraint reformulation were valid, the paper would make a useful contribution by introducing PASS to symbiotic radio and by providing two practically oriented optimization algorithms. The simulations compare against external baselines, so the performance gains are not fitted to a target result. However, the main advertised guarantee—that the design satisfies the IoT receiver's detection error probability constraint—rests on an inequality whose direction is reversed. This affects every optimized solution and the interpretation of all numerical results. The core technical claim is therefore not established, and the significance of the paper is currently conditional on a repair of the constraint derivation.","major_comments":[{"comment":"The inequality in Eq. (10) is used in the wrong direction. The cited result [37] (and the standard Pinsker-type bound for binary hypothesis testing) gives P_FA + P_MD >= 1 - sqrt(D(P0||P1)/2), i.e., Pe >= 1 - sqrt(D/2), not Pe <= 1 - sqrt(D/2). The manuscript explicitly calls Eq. (10) a 'lower bound' while writing it as an upper bound, which is internally inconsistent. Consequently, the reformulated constraint D(P0||P1) >= 2ε^2 does not imply Pe <= 1 - ε; at best it implies a lower bound that is trivially below the target for large D. Since constraint (11c) is the only mathematical encoding of the IoT receiver's detection requirement, both algorithms (LGD and SCA-PSO) solve a problem with an unverified constraint, and the abstract and conclusion's claim that the design satisfies the detection error probability constraint is unsupported. This is a load-bearing error: the detection probabi","section":"Section II-B, Eq. (10) and constraint (11c)"},{"comment":"The monotonicity claim in Eq. (33), inequality (a), is not justified. Even if problem (P2.2) is solved optimally, it is a convex surrogate of (P2.1), not the original objective. A first-order Taylor lower bound is tight only at the expansion point under certain conditions; maximizing such a surrogate does not by itself guarantee that the original objective f(W,X) does not decrease, unless the surrogate is a global lower bound that equals the original objective at the current point and the solution is chosen to dominate the current point. The paper does not provide such an argument. The convergence analysis therefore overstates the monotonic improvement of the SCA stage. This does not affect the detection-constraint error above, but it is a second load-bearing gap in the claimed convergence behavior.","section":"Section IV-C, Eq. (33)"},{"comment":"The reformulated problem in Eq. (26) is written as 'max_X' but the optimization variable at this stage is the transmit beamforming matrix W, not X. This appears to be a typographical error, but it obscures the algorithm description. Additionally, equation numbers (26), (26b) are repeated with those in (26) and (26b) inside the same display, which complicates referencing. These are presentation issues, but the first one is in a central derivation and should be corrected.","section":"Section IV-A, Eq. (26)"}],"minor_comments":[{"comment":"The definition of u2,k is misprinted: 'v1,k = |(heq_k + f_b,k)w_k|^2' should presumably be 'u2,k = ...'. The current text defines v1,k twice.","section":"Section IV-A, Eq. (24)"},{"comment":"In the simulation setup, 'the ma transmit power' appears to be a truncated phrase; also, the text uses 'Pmax/δ_k^2' to define transmit SNR but the simulation varies Pmax in dBm. Clarify the SNR definition.","section":"Section V"},{"comment":"The penalty factor μ is introduced as 'sufficiently large' and set to 10 in the simulations, but no sensitivity analysis or discussion of how μ interacts with the PSO fitness is provided. Since feasibility depends on this parameter, a brief discussion or ablation would be helpful.","section":"Section IV-B, Algorithm 2"},{"comment":"Reference [44] is cited for PSO, but the listed citation (Das and Suganthan, 'Differential evolution: A survey...') is about differential evolution, not particle swarm optimization. The PSO description should cite a standard PSO reference.","section":"References"},{"comment":"Figure 8 reports bar values but no confidence intervals, despite the results being averages over 100 channel realizations. Given that the paper's performance claims are central, reporting confidence intervals or standard deviations would strengthen the numerical evidence.","section":"Figures 8 and general reporting"}],"recommendation":"reject","confidential_remarks":"The fundamental flaw is the reversed KL bound in Eq. (10), which directly invalidates the advertised detection constraint and hence the central contribution. This is not a matter of presentation or local tuning; the constraint itself does not enforce the claimed detection error probability. The authors would need to re-derive a valid constraint or a different guarantee, which is beyond a routine revision. The paper also has useful system modeling and algorithmic ideas, so a future revision that repairs the detection constraint could be reconsidered."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, here’s my take on arXiv:2508.07002.\n\nThe new combination is real: this is the first application of pinching-antenna systems to symbiotic radio, with a joint transmit and pinching beamforming problem under a detection-error constraint. The authors propose two complementary algorithms — a learning-aided gradient descent and an SCA-PSO hybrid — and compare against sensible baselines including massive MIMO. That is a legitimate niche contribution.\n\nBut the load-bearing detection constraint is wrong. Eq. (10) says “a tractable lower bound on P_e is obtained” and then writes P_e ≤ 1 − sqrt(1/2 D(P0||P1)). The cited work (Bash et al.) gives the opposite direction: P_e ≥ 1 − sqrt(1/2 D(P0||P1)). Since P_e is false-alarm plus miss-detection probability, the correct inequality is a lower bound. The paper’s own wording contradicts the inequality symbol. So the constraint D ≥ 2ε² does not certify P_e ≤ 1−ε; it yields no upper bound at all. This is not a cosmetic typo: the abstract and conclusion claim the design “satisfies the IoT receiver’s detection error probability constraint,” and that claim rests entirely on this inequality. The simulation results in Figs. 5–9 are generated under a constraint whose advertised interpretation is unsupported.\n\nSecondary issues are less severe. The SCA convergence argument around Eq. (33) is hand-wavy; solving a convex surrogate optimally doesn’t automatically produce monotone increase in the original nonconvex objective without a proper descent lemma. That is repairable but should be tightened. The simulations lack error bars and some PSO hyperparameters are underspecified; those are minor.\n\nWhat I’d credit: the problem formulation is thoughtful, the constraint handling for antenna spacing is neat, and the complexity analysis is useful. The authors engage honestly with the prior PASS and SR literature.\n\nThis paper deserves a serious referee because the new setting is timely and the core error is identifiable and possibly fixable. But as written it should not be published. My recommendation: send to peer review, expecting major revision — correct Eq. (10), re-derive (11c), re-run the numerical study under the correct constraint, and rewrite the convergence claim. If the corrected constraint changes the optimization landscape, the performance gains may shift, so I’d want to see the new curves before trusting the conclusions.","headline":"The PASS–SR combination is new and the algorithms are sensible, but Eq. (10) reverses the direction of the KL bound, so the advertised detection guarantee and all constrained results rest on an invalid inequality.","tokens_in":19797,"tokens_out":3300,"would_cite":false,"duration_ms":31170,"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":"Jointly optimizing transmit beamforming and pinching-antenna positions in a symbiotic radio downlink raises sum rate by up to 35.5% over fixed placement.","keywords":["symbiotic radio","pinching antenna system","beamforming optimization","backscatter communication","successive convex approximation","particle swarm optimization","detection error probability","sum-rate maximization"],"falsifier":"Compute the exact minimum detection error probability $P_e$ for the OOK hypotheses in Eq. (7) by numerical integration over the likelihood ratio, and compare it with $1-\\sqrt{D(P_0\\|P_1)/2}$ across the system's operating SNRs. If the exact $P_e$ does not stay at or below that bound, or if the source cited for Eq. (10) gives the opposite inequality, then constraint (11c) does not guarantee the stated detection performance.","tokens_in":18762,"feed_emoji":"📡","tokens_out":6542,"duration_ms":74567,"temperature":0.7,"pith_summary":"This paper proposes a downlink symbiotic radio system in which a base station uses pinching antennas—small radiating elements that slide along dielectric waveguides—to serve primary receivers while a backscatter device carries IoT data to an IoT receiver. The authors claim that jointly optimizing the transmit beamforming matrix and the positions of the pinching antennas, subject to a detection-error-probability constraint on the IoT link, substantially increases the achievable sum rate. Two solution methods are developed: a learning-aided gradient descent approach that treats the variables as trainable parameters, and a two-stage successive convex approximation plus particle swarm optimization approach. Their simulations place the two-stage method 17.1% above the gradient-descent method and 35.5% above a fixed placement, and close to an element-wise exhaustive search.","feed_headline":"Pinching antennas lift symbiotic-radio sum rate by 35.5%","feed_subtitle":"Jointly re-aiming transmit beams and movable antenna positions beats fixed placement and closes in on exhaustive search.","key_machinery":"The pinching antenna system (PASS): dielectric waveguides populated by movable pinching antennas whose positions $x_{n,m}$ enter the channel as both phase shifts and distance-dependent path loss. The load-bearing mechanism is the joint optimization of the transmit beamforming matrix $\\mathbf{W}$ and the position matrix $\\mathbf{X}$, with positions reparameterized as non-negative offsets $\\Delta x_{n,m}$ to enforce minimum-spacing constraints, and the detection requirement expressed through the KL divergence $D(P_0\\|P_1)$.","core_discovery":"The central claim is that reconfigurable pinching-antenna placement turns large-scale path loss and signal phase into controllable design variables, so a pinching-antenna base station can create strong line-of-sight links to primary receivers while managing the backscatter link for the IoT receiver. The paper formulates a sum-rate maximization over the transmit beamforming matrix W and the pinching position matrix X, constrained by transmit power, minimum antenna spacing, waveguide length, and a detection-error-probability requirement on the IoT receiver. It then solves this coupled nonconvex problem two ways—a differentiable reparameterization solved by gradient descent, and an alternating","pith_inferences":["The offset reparameterization used here transfers directly to any movable-antenna problem with minimum-spacing constraints, so it could be reused beyond symbiotic radio.","Because the simulations use a line-of-sight-only channel model, the reported gains likely represent the upper end of what pinching antennas can offer; a rich-scattering environment would be a direct stress test.","The authors point to parasitic (same-symbol-duration) symbiotic radio as future work; extending the framework there would test whether the sum-rate gains survive under tighter timing constraints."],"forward_implications":["If the central claim holds, pinching antennas give symbiotic radio a physical way to mitigate the weak double-fading backscatter link by moving radiators, not just by precoding.","Joint position and beam optimization remains effective under a strict detection-error constraint on the IoT receiver, broadening where backscatter IoT services can be deployed.","The two-stage SCA-PSO method achieves near-element-wise performance at much lower complexity, making it a practical candidate for implementation.","The reported gains widen with higher transmit SNR and larger waveguide range, suggesting the advantage is tied to path-loss and phase control.","The learning-aided gradient descent method offers a low-complexity plug-and-play alternative, though it is more prone to local optima."],"supporting_citations":[{"why":"Defines the pinching-antenna waveguide concept on which the entire system is built.","marker":"[19]"},{"why":"Supplies the flexible-antenna channel model and low-complexity pinching beamforming that this paper extends to symbiotic radio.","marker":"[20]"},{"why":"Provides the geometric line-of-sight channel model and the element-wise position-search benchmark used in simulations.","marker":"[22]"},{"why":"Establishes the joint transmit-and-pinching beamforming problem formulation that this paper adapts to the SR setting.","marker":"[25]"},{"why":"Defines symbiotic radio and the primary/secondary transmission relationship being optimized.","marker":"[4]"},{"why":"Derives the detection error probability expression and likelihood-ratio test behind the IoT receiver constraint.","marker":"[36]"},{"why":"Gives the KL-divergence inequality on which the detection-error constraint (11c) is based.","marker":"[37]"},{"why":"Supplies the numerical system parameters used in the performance evaluations.","marker":"[48]"}],"fun_headline_variants":["Joint transmit and pinching beamforming boosts symbiotic radio rates","Movable pinching antennas optimize joint beamforming for higher sum rate","Adaptive pinching antenna positions enhance symbiotic radio sum rate","Joint beamforming with reconfigurable pinching antennas lifts rates","Pinching antenna positions and transmit beamforming jointly optimized"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The load-bearing premise is that the inequality $P_e \\le 1-\\sqrt{D(P_0\\|P_1)/2}$ in Eq. (10) has the correct direction, so requiring $D(P_0\\|P_1)\\ge 2\\varepsilon^2$ truly enforces $P_e \\le 1-\\varepsilon$ on the IoT receiver's detection error.","fun_headline_variants_meta":{"raw":{"variants":["Joint transmit and pinching beamforming boosts symbiotic radio rates","Movable pinching antennas optimize joint beamforming for higher sum rate","Adaptive pinching antenna positions enhance symbiotic radio sum rate","Joint beamforming with reconfigurable pinching antennas lifts rates","Pinching antenna positions and transmit beamforming jointly optimized"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000505,"raw_usage":{"total_tokens":2314,"prompt_tokens":768,"completion_tokens":1546,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":512,"completion_tokens_details":{"reasoning_tokens":1463}},"tokens_in":512,"tokens_out":1546,"duration_ms":14301,"temperature":1.0,"reasoning_tokens":1463,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T22:25:11.109184+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the exact minimum detection error probability $P_e$ for the OOK hypotheses in Eq. (7) by numerical integration over the likelihood ratio, and compare it with $1-\\sqrt{D(P_0\\|P_1)/2}$ across the system's operating SNRs. If the exact $P_e$ does not stay at or below that bound, or if the source cited for Eq. (10) gives the opposite inequality, then constraint (11c) does not guarantee the stated detection performance.","supporting_citations":[],"review_version":1}