{"id":"333d62c1-0100-4f3f-aef9-25f824304d60","arxiv_id":"2509.03038","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The optimal design for a single-user pinching-antenna SWIPT link under random blockage is to place the antenna at the closest feasible point to the user and set the split to exactly meet the harvesting target.","lead":"This paper derives a closed-form rule for placing a movable pinching antenna and setting the power-splitting ratio in a wireless link that sends both data and power to a single user while obstacles randomly block the direct path. The rule is to put the antenna at the feasible point closest to the user and give the energy harvester just enough power to meet its target.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1's boundary handling is flawed: R=0 can be feasible with rho=0, and clip is undefined when the feasible interval misses [0,L].","rationale":"The paper's central claim is the closed-form global optimality in Theorem 1. The most load-bearing concern is that Theorem 1 is not correct as stated in boundary cases: it declares R=0 infeasible when a feasible rho=0 solution exists, and it applies a clip formula that is undefined when the feasible interval does not intersect the waveguide span. These are internal inconsistencies that directly affect the claimed universal closed-form solution. The reader's rationale explicitly noted both boundary issues, so this is not a new concern. However, the reader's 'weakest_assumption' focused on the channel model (exponential blockage, negligible NLoS). That is a valid external-validity concern, but it is a modeling assumption rather than a proof error; the theorem is conditional on that model. The boundary error is more load-bearing because it invalidates the theorem within the paper's own framework and would mislead any implementation of the closed form. The core monotonicity and Lambert W derivation are correct for the typical regime R>0 with nonempty intersection, so the paper's main insight survives after adding a feasibility check and fixing the R=0 branch; hence the verdict remains CONDITIONAL rather than REJECT. I agree with the reader's conditional recommendation but would prioritize the boundary correction over the channel-model robustness concern.","tokens_in":8655,"tokens_out":9279,"duration_ms":81274,"concrete_test":"Set beta=0.01, eta=(c/(4*pi*fc))^2 at 28 GHz, zeta=0.6, Pt=10 W, y_u=5, z_p=10, xu=5, L=50, and choose q0 = zeta*Pt*eta*exp(-beta*125)/125 (approx 1e-8 W). Then t_th = 125 = y_u^2+z_p^2, so R=0. Evaluating the theorem's decision rule returns 'no feasible solution', but direct substitution x=5, rho=0 gives zeta(1-rho)Pt f(5) = q0 and SNR = 0, so P0 is feasible. Additionally, with xu=200 and q0 chosen so t_th=200, R=sqrt(75)>0 but [xu-R,xu+R] is disjoint from [0,L]; applying the clip formula by clamping returns x=50, which does not satisfy (50-200)^2 + 125 <= 200. Run both cases to confirm the theorem's output is incorrect or undefined.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1 states that if R=0 no feasible solution exists, but this contradicts Lemma 1 and the paper's own formulation. When t_th = y_u^2 + z_p^2, the feasibility condition f(x) >= q0/(zeta Pt) holds exactly at x = xu (if xu in [0,L]), with rho = 0 and SNR = 0. This is a feasible point, so the 'no feasible solution' branch is false. The theorem also gives x* = clip(xu, max{0,xu-R}, min{L,xu+R}) without checking whether [xu-R, xu+R] intersects [0,L]. If it does not, the interval for clip is empty; a clamping implementation would silently return an infeasible x (e.g., xu far outside [0,L] with R>0 but (proj_{[0,L]}(xu)-xu)^2 > R^2). Thus the closed-form is not universally valid: it requires an explicit feasibility test (min_{x in [0,L]} (x-xu)^2 <= R^2) and a correct R=0 case. This is an internal inconsistency in the central claim, not merely a matter of model realism.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies a single-user downlink SWIPT system with a movable pinching antenna over a waveguide, under a probabilistic LoS blockage model where Pr(LoS)=exp(-β d^2(x)) and non-LoS contributions are neglected. The effective complex channel is h=γ h_LoS, and the average channel power becomes f(x)=η exp(-β d^2(x))/d^2(x). The authors formulate a joint optimization of the PA position x and the power-splitting ratio ρ to maximize average SNR subject to an average harvested-power constraint q0 and the placement constraint 0≤x≤L. They prove that f(x) is strictly decreasing in the squared distance d^2(x), reduce the problem to maximizing f(x) over an EH-feasible interval, and derive a closed-form solution involving a Lambert W threshold t_th and a projection/clip operation. Numerical grid-search results are reported to validate the closed form in the simulated interior cases.","tokens_in":8828,"tokens_out":7804,"duration_ms":75313,"significance":"If the boundary cases are repaired, the result is a genuinely useful exact solution to a non-convex joint antenna-placement and power-splitting problem: the design reduces to one Lambert W evaluation and one projection. The monotonicity argument, the Lambert W feasibility threshold, and the optimality reduction to maximizing f(x) are all derived correctly and are presented concisely. The numerical validation supports the main claim for the non-degenerate parameter regimes considered. Since the result is closed-form and globally optimal within the stated model, it provides a clean design insight for reconfigurable-antenna SWIPT and should be of interest to the community, provided the theorem statement and feasibility handling are corrected and the scope-limiting assumptions are made explicit.","major_comments":[{"comment":"The R=0 branch is incorrect. If t_th = y_u^2 + z_p^2, then R=0, but x=x_u is feasible whenever x_u∈[0,L]: at that point f(x_u)=q0/(ζP_t) by the definition of t_th, so the EH constraint holds with equality at ρ=0, giving SNR=0. This feasible point contradicts the theorem's claim that 'no feasible solution exists' for R=0. The problem is that R is defined as sqrt(max{0, t_th-(y_u^2+z_p^2)}), which erases the sign of D=t_th-(y_u^2+z_p^2); D<0 is infeasible, but D=0 with x_u∈[0,L] is feasible exactly at x=x_u with ρ=0. The theorem must restate the boundary case separately, e.g. by returning x*=x_u, ρ*=0 when D=0 and x_u∈[0,L], and infeasibility otherwise.","section":"Theorem 1, Eq. (12)"},{"comment":"The closed-form x* formula assumes that the feasible interval [x_u-R, x_u+R] ∩ [0,L] is nonempty, but no such check appears in the theorem. If dist(x_u,[0,L])^2 > R^2, then the interval is empty, the lower bound max{0,x_u-R} can exceed the upper bound min{L,x_u+R}, and the clip operation is undefined; an implementation would silently return an infeasible point. For example, with L=50, x_u=100, and R=1, the feasible interval misses [0,L] entirely even though R>0. The theorem should include the explicit feasibility condition min_{x∈[0,L]} (x-x_u)^2 ≤ R^2 and return 'no feasible solution' when it fails. With that condition added, the projection formula is correct in the nonempty case.","section":"Eq. (20) and proof after Eq. (18)"},{"comment":"The derivation of the closed form is contingent on the average channel power being exactly f(x)=η exp(-β d^2(x))/d^2(x), which follows only because non-LoS components are assumed negligible and the LoS probability is assumed to be exp(-β d^2(x)). This is stated in the model paragraph after Eq. (3), but it is load-bearing for both the strict monotonicity of f(x) in d^2(x) and the Lambert W feasibility threshold. The paper should state prominently, for example in a remark following Theorem 1, that the global-optimality claim is conditional on this model; if NLoS components were non-negligible or the blockage probability had a different distance dependence, the closed form and the threshold would not apply. This is a scope limitation rather than an internal inconsistency, but it deserves more emphasis given the strength of the claimed global optimality.","section":"Section II-B, Eqs. (2)-(3)"}],"minor_comments":[{"comment":"Appendix A is titled 'PROOF OF PROPOSITION 1' but contains no text, and there is no Proposition 1 in the paper. Either remove the appendix or supply the missing content.","section":"Appendix A"},{"comment":"The paragraph beginning 'Fig. 3(a) presents the average SNR versus the transmit power Pt' appears twice, with slightly different wording; one duplicate should be deleted. The figure references in the two versions also conflict.","section":"Section IV"},{"comment":"The piecewise display for ρ* is malformed: for the R>0 case the expression for x* is shown before the brace for ρ*, and the R=0 case is handled separately. After the boundary-case correction, the theorem should present a single well-formed case structure for both x* and ρ*.","section":"Eq. (12)"},{"comment":"The sentence justifying the omission of in-waveguide attenuation cites references [15] and [16]; one of these is co-authored by a current author of this manuscript. This is not a technical problem, but the paper may wish to add a sentence clarifying the basis for the omission beyond the cited analysis.","section":"Section II-B, paragraph after Eq. (3)"}],"recommendation":"major_revision","confidential_remarks":"The two technical flaws in Theorem 1 are localized and fixable within the manuscript's scope: the R=0 boundary case and the missing feasibility check for the intersection of the EH-feasible interval with [0,L]. Once corrected, the central derivation is sound and the contribution is a useful closed-form result. I also recommend that the authors explicitly acknowledge the model-dependence of the monotonicity argument, as overclaiming global optimality beyond the assumed blockage statistics would be easy for readers to do. The self-citation for the in-waveguide attenuation assumption is appropriate and not a concern."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, the paper's main result is a genuinely useful closed form: for a single pinching antenna serving one SWIPT user under probabilistic LoS blockage, the optimal position is the point in the EH-feasible interval closest to the user, and the optimal PS ratio is the smallest needed to meet the average harvested-power target. The monotonicity argument and the Lambert W threshold are derived correctly, and the simulations confirm the formula. Second, Theorem 1 as stated has a boundary-case error. It declares R=0 infeasible, but when t_th = y_u^2+z_p^2 and xu∈[0,L], x=xu with rho=0 is feasible (EH constraint holds with equality). The clip expression also silently fails when the interval [xu-R, xu+R] does not intersect [0,L]; the paper needs an explicit feasibility test, e.g., min_{x∈[0,L]} (x-xu)^2 ≤ R^2, and a separate handling of the equality case. These are fixable, but they are inside the central theorem, so they should be corrected before this is published.\n\nWhat the paper does well: it is the first in its cited literature to combine PA position and PS ratio under random LoS blockage, and the closed form is simple enough to be a design rule. The derivation is self-contained: no parameters are fitted to the result. The channel model is a strong idealization—exponential blockage, no NLoS, no in-waveguide loss—but those assumptions are stated plainly, not hidden, and they are consistent with recent PA papers.\n\nThe manuscript also needs an editorial pass: Section II contains duplicated text blocks, and there is an 'Appendix A: Proof of Proposition 1' that refers to no Proposition 1. These are cosmetic but show the version is rough.\n\nWho this is for: researchers working on pinching-antenna design, SWIPT with stochastic channels, and reconfigurable antennas generally. They will get a clean rule of thumb and a clear feasibility threshold. The paper deserves a serious referee: the contribution is real, and the flaws are precisely locatable. My recommendation is to send it to peer review with a request to fix the R=0 branch and add the feasibility test, then it can be accepted.","headline":"A useful closed-form design rule for PA-assisted SWIPT under probabilistic LoS blockage, marred by a real but fixable boundary-case error in Theorem 1.","tokens_in":9423,"tokens_out":3420,"would_cite":true,"duration_ms":29444,"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":"For a pinching-antenna SWIPT system under probabilistic LoS blockage, the jointly optimal antenna position and power-splitting ratio are given in closed form: place the antenna closest to the user's coordinate that satisfies the…","keywords":["simultaneous wireless information and power transfer","pinching antenna","probabilistic line-of-sight blockage","power splitting","energy harvesting","Lambert W function","closed-form optimization"],"falsifier":"Measure the average channel power $\\mathbb{E}[|h|^2]$ for a PA sliding along the waveguide in a cluttered environment and plot it against squared PA-user distance. The closed form requires this curve to be strictly decreasing; any observed local maximum away from the user, or a nonzero floor from NLoS paths at large distance, falsifies the monotonicity premise and with it the optimality of the clip solution. A numerical version: re-solve the problem with the blockage probability $\\Pr(\\gamma=1)=\\exp(-\\beta d)$ instead of $\\exp(-\\beta d^2)$; the resulting optimal position generally will not match the paper's formula.","tokens_in":8419,"feed_emoji":"⚡","tokens_out":13424,"duration_ms":116362,"temperature":0.7,"pith_summary":"This paper studies a single user receiving data and harvesting energy from a movable pinching antenna sliding along a waveguide, in an environment where the line-of-sight path is blocked with probability that grows with distance. It claims that the joint choice of antenna position and power-splitting ratio has a closed-form global optimum: the antenna should be placed as close as possible to the user's horizontal coordinate, and the splitter should send to energy harvesting only the minimum power needed to meet the harvesting target, with the rest going to information decoding. The energy target therefore enters mainly through a feasibility threshold, expressible with the Lambert W function, rather than through a trade-off that requires numerical search. If correct, this reduces a non-convex resource allocation problem to one projection and one Lambert W evaluation, and it gives a concrete rule for when mechanical antenna reconfigurability makes a SWIPT link sustainable under random blockage.","feed_headline":"One formula fixes antenna position and power split under blockage","feed_subtitle":"A Lambert W check decides whether energy harvesting is feasible; if so, the antenna stays at the user's coordinate.","key_machinery":"The load-bearing object is the average channel power $f(x)=\\eta\\exp(-\\beta d^2(x))/d^2(x)$, with $d^2(x)=(x-x_u)^2+y_u^2+z_p^2$. Its strict monotonicity in the squared distance, $g'(t)=-e^{-\\beta t}(\\beta t+1)/t^2<0$ for all $t>0$, makes maximizing average SNR equivalent to minimizing distance to the user, so the antenna-position problem becomes a projection. The second piece is the Lambert W function, the inverse of $w\\mapsto w e^w$; it converts the feasibility inequality $t e^{\\beta t}\\le \\zeta P_t\\eta/q_0$ into the explicit threshold $t_{\\mathrm{th}}=\\frac{1}{\\beta}W(\\beta\\eta\\zeta P_t/q_0)$. Together these turn a non-convex joint optimization into a one-line closed form: define the feasible radius $R$ from $t_{\\mathrm{th}}$ minus the fixed vertical and lateral offsets, intersect the interval $[x_u-R,x_u+R]$ with $[0,L]$, and project $x_u$ onto that intersection.","core_discovery":"The paper's central claim is Theorem 1: for the power-splitting SWIPT system with a pinching antenna and probabilistic LoS blockage, the optimal PA position and PS ratio are $x^\\star = \\operatorname{clip}(x_u,\\max\\{0,x_u-R\\},\\min\\{L,x_u+R\\})$ and $\\rho^\\star = 1 - q_0/(\\zeta P_t f(x^\\star))$, where $R = \\sqrt{\\max\\{0,t_{\\mathrm{th}}-(y_u^2+z_p^2)\\}}$ and $t_{\\mathrm{th}} = \\frac{1}{\\beta}W\\!\\left(\\frac{\\beta\\eta\\zeta P_t}{q_0}\\right)$. The derivation first decouples the problem: for any fixed $x$, the EH constraint forces $\\rho \\le 1 - q_0/(\\zeta P_t f(x))$, and since SNR is increasing in $\\rho$, the optimal split uses equality, leaving SNR proportional to $f(x) - q_0/(\\zeta P_t)$. Maximizing $f(x) = \\eta e^{-\\beta d^2(x)}/d^2(x)$ over $x\\in[0,L]$ then reduces to minimizing squared distance $d^2$, because $g(t)=e^{-\\beta t}/t$ is strictly decreasing for $t>0$. The EH constraint becomes the quadratic feasibility condition $t\\le t_{\\mathrm{th}}$, which the Lambert W function solves exactly. Consequently the global optimum is the point closest to the user inside the EH-feasible interval, with $\\rho^\\star$ reserving just enough power for harvesting.","pith_inferences":["A direct test of the claimed structure: fix the user's $x$-coordinate strictly inside the waveguide and vary the harvesting target; Theorem 1 predicts the optimal antenna position stays at $x_u$ until the problem becomes infeasible, while $\\rho^\\star$ decreases continuously. Measuring only the antenna position would incorrectly suggest the energy constraint is irrelevant; the real signature is the","The closed form is specific to the quadratic exponent in the blockage law. For a blockage probability of the form $\\exp(-\\beta d^\\alpha)$ with $\\alpha\\neq 2$, the monotonicity of the average gain survives, so the optimal-position-as-projection structure is preserved, but the feasibility threshold solves $t\\exp(\\beta t^{\\alpha/2}) = \\zeta P_t\\eta/q_0$ numerically rather than through Lambert W.","In a nonlinear energy-harvesting model where harvested DC power is a concave function of received RF power, Lemma 1's step of setting $\\rho$ to its maximum feasible value would no longer be optimal; the decoupling and the closed form would need re-derivation."],"forward_implications":["The joint problem decouples exactly: at the optimum the power splitter reserves only the minimum received power needed for the harvesting target, $\\rho^\\star=1-q_0/(\\zeta P_t f(x^\\star))$, and the maximum average SNR equals $\\lambda(f(x^\\star)-q_0/(\\zeta P_t))$, a fixed SNR penalty $-\\lambda q_0/(\\zeta P_t)$ relative to the no-harvesting case.","Feasibility is a closed-form condition: the system can meet the harvesting target if and only if $\\frac{1}{\\beta}W(\\beta\\eta\\zeta P_t/q_0)\\ge y_u^2+z_p^2$; otherwise no antenna position or split ratio works.","When the user's horizontal coordinate lies on the waveguide, the optimal antenna position is exactly that coordinate whenever any feasible position exists; the harvesting requirement changes $\\rho^\\star$ and the size of the feasible region, but does not move the antenna away from the user.","Increasing blockage density $\\beta$ or harvesting target $q_0$ shrinks the feasible radius $R$ and can make the problem infeasible; within the feasible regime, the split ratio $\\rho^\\star$ falls (more power to harvesting) as $\\beta$ or $q_0$ grows.","Fixed-antenna or fixed-split designs are strictly suboptimal because they either forgo the average channel gain of the best position or waste received power on harvesting; the gain of the closed-form scheme is largest where the EH constraint is tight."],"supporting_citations":[{"why":"establishes the pinching-antenna mechanism and its mechanical reconfigurability, which the system model assumes.","marker":"[3]"},{"why":"prior PA-aided SWIPT optimization showing joint PA positioning and power design, the line of work this paper extends to blockage.","marker":"[10]"},{"why":"prior PA-assisted SWIPT framework that this paper extends to probabilistic LoS blockage.","marker":"[13]"},{"why":"source for the claim that probabilistic LoS blockage changes PA-system behavior, grounding the Bernoulli blockage model.","marker":"[14]"},{"why":"supplies the probabilistic LoS blockage model and the analytical justification for ignoring in-waveguide attenuation.","marker":"[15]"},{"why":"supports the omission of in-waveguide attenuation in the channel model.","marker":"[16]"}],"fun_headline_variants":["One formula pinpoints antenna and power split under shadows","Closed-form answer for antenna placement under LoS blockage","Lambert W governs SWIPT antenna position and energy split","Optimal pinching antenna: a single equation does it","Blockage? Just solve for antenna spot and power ratio"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire closed form rests on the effective channel being only the LoS path, with non-LoS contributions negligible and the LoS blockage probability equal to $\\exp(-\\beta$ times squared distance$)$; if reflections add a meaningful floor to the average channel gain or the blockage law has a different distance dependence, the strict monotonicity that pins the antenna to the user's coordinate can fail.","fun_headline_variants_meta":{"raw":{"variants":["One formula pinpoints antenna and power split under shadows","Closed-form answer for antenna placement under LoS blockage","Lambert W governs SWIPT antenna position and energy split","Optimal pinching antenna: a single equation does it","Blockage? Just solve for antenna spot and power ratio"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000244,"raw_usage":{"total_tokens":1580,"prompt_tokens":1040,"completion_tokens":540,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":656,"completion_tokens_details":{"reasoning_tokens":461}},"tokens_in":656,"tokens_out":540,"duration_ms":5766,"temperature":1.0,"reasoning_tokens":461,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T16:33:36.025897+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the average channel power $\\mathbb{E}[|h|^2]$ for a PA sliding along the waveguide in a cluttered environment and plot it against squared PA-user distance. The closed form requires this curve to be strictly decreasing; any observed local maximum away from the user, or a nonzero floor from NLoS paths at large distance, falsifies the monotonicity premise and with it the optimality of the clip solution. A numerical version: re-solve the problem with the blockage probability $\\Pr(\\gamma=1)=\\exp(-\\beta d)$ instead of $\\exp(-\\beta d^2)$; the resulting optimal position generally will not match the paper's formula.","supporting_citations":[{"cited_title":"MIMO Pinching-Antenna-Aided SWIPT","cited_arxiv_id":"2506.06754","evidence_quote":"prior PA-aided SWIPT optimization showing joint PA positioning and power design, the line of work this paper extends to blockage."},{"cited_title":"LoS blockage in pinching-antenna systems: Curse or blessing?","cited_arxiv_id":null,"evidence_quote":"source for the claim that probabilistic LoS blockage changes PA-system behavior, grounding the Bernoulli blockage model."}],"review_version":1}