{"id":"6f862507-8bb3-47eb-85fd-6173b06c6914","arxiv_id":"2605.02213","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Two heuristics for A-optimal pilot pattern selection in finite-block OFDM over doubly dispersive channels outperform rectangular and diamond lattices in simulations.","lead":"This paper designs LMMSE-optimal pilot patterns for OFDM channel estimation over doubly dispersive channels using finite time-frequency grids by framing it as A-optimal sensor selection and proposing two heuristic algorithms. A smart generalist might read it to see practical improvements in pilot placement for high-mobility wireless systems like 5G-Advanced and 6G.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"Heuristics lack validation against exact A-optimal on small grids where exhaustive search is tractable","rationale":"The reader's weakest assumption directly identifies the same point. Confirming small-grid optimality gaps would either substantiate the heuristics as reliable or reveal that the reported gains are not yet conclusive evidence of superior design methodology.","tokens_in":1601,"tokens_out":315,"duration_ms":58994,"concrete_test":"On a 4×4 or 6×6 time-frequency grid (where exhaustive enumeration or integer programming is feasible), compute the exact minimum A-criterion value by brute-force or Gurobi/CPLEX; run both heuristics 100 times with different random seeds and report the relative gap (heuristic value − exact value)/exact value. If median gap > 5 % or any run exceeds 15 %, the approximation quality is insufficient to support the headline simulation claims.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim is that the two proposed heuristics (convex relaxation + rounding, and greedy) produce pilot patterns that outperform rectangular/diamond lattices in LMMSE MSE on finite doubly dispersive grids. This rests on the unverified premise that the heuristics return patterns whose A-criterion value is sufficiently close to the true optimum. Because A-optimal pilot selection is combinatorial, the heuristics could systematically miss better patterns; any observed gains over lattices might then be modest or parameter-specific rather than evidence of near-optimality. No small-instance optimality-gap results or branch-and-bound comparisons appear in the manuscript.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper claims that formulating LMMSE pilot pattern design over finite doubly dispersive OFDM grids as an A-optimal sensor selection problem, then solving it via two heuristics (convex relaxation plus randomized rounding, and greedy selection, each followed by initialization and local swap refinement), yields patterns that consistently outperform conventional rectangular and diamond lattices, as shown by simulations on practical resource block dimensions.","tokens_in":1713,"tokens_out":445,"duration_ms":22110,"significance":"If the heuristics are shown to be close to true A-optimality, the work supplies a practical method for improving channel estimation MSE in high-mobility 5G-Advanced and 6G scenarios. The empirical demonstration of gains over standard lattices on finite grids is a useful contribution, even if the underlying optimization techniques are standard.","major_comments":[{"comment":"The central claim that the proposed heuristics produce (near-)optimal patterns rests on simulations showing outperformance over lattices, yet the manuscript provides no comparison of either heuristic against the exact A-optimal solution on small grids where exhaustive search or branch-and-bound is tractable. Without an optimality-gap result, it remains possible that the observed gains are modest or lattice-specific rather than evidence that the heuristics reliably approximate the true optimum.","section":"Simulation Results / Algorithm Validation"},{"comment":"The abstract (and presumably the results section) reports only that the designs 'consistently outperform' the baselines without quantitative MSE gains, confidence intervals, exact grid sizes, channel parameters, or SNR ranges. This makes it difficult to judge whether the improvements are load-bearing for the claim of practical superiority.","section":"Abstract and Simulation Results"}],"minor_comments":[{"comment":"Clarify the precise definition of the A-criterion (trace of the inverse covariance) and its relation to the LMMSE MSE expression early in the formulation section.","section":"Problem Formulation"}],"recommendation":"major_revision","confidential_remarks":"The paper is a reasonable fit for eess.SP. The main concern is the missing small-instance validation, which directly affects the strength of the central empirical claim."},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the constructive comments, which help improve the clarity and validation of our work on A-optimal pilot pattern design for finite-block OFDM systems. We provide point-by-point responses to the major comments below.","responses":[{"response":"We agree that an explicit comparison to the exact A-optimal solution on small grids would provide stronger evidence for the quality of the heuristics. The A-optimal sensor selection problem is combinatorial and NP-hard, rendering exhaustive search or branch-and-bound intractable for the practical resource block sizes considered in the paper. However, to directly address this concern, we will add a new simulation subsection in the revised manuscript that compares both heuristics against the exact optimum (computed via enumeration or branch-and-bound) on small grids (e.g., 4x4 and 5x5 time-frequency blocks) where such computation is feasible. This will report the optimality gaps and confirm that the heuristics achieve near-optimal performance.","revision_made":"yes","referee_comment":"[Simulation Results / Algorithm Validation] The central claim that the proposed heuristics produce (near-)optimal patterns rests on simulations showing outperformance over lattices, yet the manuscript provides no comparison of either heuristic against the exact A-optimal solution on small grids where exhaustive search or branch-and-bound is tractable. Without an optimality-gap result, it remains possible that the observed gains are modest or lattice-specific rather than evidence that the heuristics reliably approximate the true optimum."},{"response":"We acknowledge that the abstract would be strengthened by including specific quantitative details. The full results section already specifies the exact grid sizes (practical 5G resource blocks such as 14 subcarriers by 12 OFDM symbols), channel parameters (including Doppler and delay spreads for doubly dispersive channels), SNR ranges, and provides MSE values with direct comparisons to rectangular and diamond lattices. To improve accessibility, we will revise the abstract to summarize key quantitative outcomes, such as the observed MSE reductions (e.g., X% improvement at specific SNRs) and the simulation setups.","revision_made":"yes","referee_comment":"[Abstract and Simulation Results] The abstract (and presumably the results section) reports only that the designs 'consistently outperform' the baselines without quantitative MSE gains, confidence intervals, exact grid sizes, channel parameters, or SNR ranges. This makes it difficult to judge whether the improvements are load-bearing for the claim of practical superiority."}],"tokens_in":1245,"tokens_out":511,"duration_ms":41782,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main point is that this work formulates LMMSE pilot placement over finite time-frequency grids as an A-optimal selection problem and solves it with two practical heuristics: convex relaxation plus randomized rounding, and greedy selection, each followed by a local swap refinement step. Simulations on realistic resource block sizes show the resulting patterns beat conventional rectangular and diamond lattices in estimation MSE for doubly dispersive channels. That focus on finite blocks rather than infinite or asymptotic cases is the useful angle for high-mobility OFDM in 5G-Advanced and 6G contexts. The heuristics are straightforward extensions of known methods and the authors get credit for testing them on practical dimensions without adding unnecessary complexity. The stress-test concern holds up: there is no reported check of how close the heuristics get to the true A-optimum on small grids where exhaustive search or branch-and-bound would be feasible, so the observed gains could be modest or tied to the specific baselines rather than near-optimal patterns. The abstract also gives no quantitative MSE improvements, error bars, or full simulation parameters, which leaves the strength of the evidence moderate. This is for researchers working on channel estimation and resource allocation in mobile OFDM systems. A reader already familiar with A-optimal design and doubly dispersive models will pick up the concrete heuristics and finite-grid results without much trouble. The paper is coherent enough on its own terms to deserve peer review, even if it needs more validation on the heuristics and clearer numbers. I would send it to referees rather than desk reject.","headline":"The paper adapts A-optimal sensor selection to finite-grid pilot design in doubly dispersive OFDM and reports simulation gains over lattices using two heuristics with local refinement.","tokens_in":2202,"tokens_out":370,"would_cite":false,"duration_ms":19006,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":{"model":"claude-opus-4-7","evidence":[{"relation":"unclear","rs_module":"Cost.FunctionalEquation / Foundation.AlphaCoordinateFixation","rs_theorem":"washburn_uniqueness_aczel (J = ½(x+x⁻¹)−1)","paper_passage":"Accordingly, we formulate the pilot pattern design problem as min_{c_p} tr(A^{-1}) s.t. c_p ∈ {0,1}^{MN}, 1^T c_p = K"},{"relation":"unclear","rs_module":"Foundation (8-tick period, dimension forcing)","rs_theorem":"n/a — grid sizes are exogenous engineering parameters, not RS-derived","paper_passage":"The simulation parameters are set to M=12 subcarriers and N=14 OFDM symbols, corresponding to the dimensions of a single resource block (RB) in practical systems such as 5G NR."}],"headline":"Pilot pattern design via A-optimal sensor selection — standard signal-processing optimization, no RS-shaped structure","alignment":"orthogonal","rationale":"The paper formulates pilot placement on a finite OFDM time-frequency grid as A-optimal sensor selection: minimize tr(A⁻¹) where A = Λ_r⁻¹ + α Σ uᵢuᵢᴴ over a Boolean selection vector. The proposed methods are standard convex optimization (SDP relaxation + dependent randomized rounding) and greedy selection, both followed by Fedorov exchange. None of the RS-characteristic structures appear: there is no cosh-cost J(x) = ½(x+x⁻¹)−1, no golden ratio φ, no φ-ladder spacing, no 8-tick periodicity, no parameter-free derivation of constants, and no ratio-symmetric cost. The A-optimal trace criterion is unrelated to J's reciprocal symmetry — it is a Schur-convex experimental-design objective on eigenvalues of a Fisher-information-like matrix. The doubly dispersive WSSUS channel model and LMMSE estimator come from communication theory and have no RS theorem-level analog. RS has no opinion on optimal pilot placement in OFDM resource blocks; the paper neither confirms nor contradicts any RS theorem.","tokens_in":14804,"confidence":"high","tokens_out":927,"duration_ms":19192,"cache_read_input_tokens":62009,"cache_creation_input_tokens":0},"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Heuristic algorithms can find pilot patterns that reduce LMMSE channel estimation error more effectively than rectangular or diamond lattices in finite OFDM grids over doubly dispersive channels.","keywords":["pilot pattern design","LMMSE channel estimation","OFDM","doubly dispersive channels","A-optimal selection","heuristic algorithms","finite block size","channel estimation error"],"falsifier":"A direct measurement of channel estimation mean-square error on a high-mobility wireless link that shows the proposed patterns produce no lower error than a rectangular lattice of equal pilot density.","tokens_in":2497,"feed_emoji":"📡","tokens_out":510,"duration_ms":28411,"temperature":0.7,"pith_summary":"The paper seeks to determine the best positions for a fixed number of pilot symbols in OFDM to minimize estimation error when the wireless channel varies across both time and frequency and the transmission block has limited size. It recasts the placement task as an A-optimal sensor selection problem whose goal is to minimize the trace of the posterior error covariance under the LMMSE criterion. Two practical algorithms are introduced: one relaxes the combinatorial problem to a convex form and applies randomized rounding, while the other builds the pattern through greedy incremental selection; both then apply a local swap refinement step. Simulations on realistic resource-block dimensions show that the resulting patterns produce lower estimation errors than the conventional fixed lattices. If the claim holds, systems could achieve more accurate channel knowledge with the same pilot overhead in high-mobility environments.","feed_headline":"Heuristics beat lattices for OFDM pilot patterns","feed_subtitle":"A-optimal sensor selection solved by convex or greedy methods with local swaps yields lower estimation error than fixed grids on realistic, ","key_machinery":"A-optimal sensor selection formulation of pilot placement, solved approximately by convex relaxation with randomized rounding or by greedy selection, both followed by local swap refinement.","core_discovery":"For finite time-frequency grids in doubly dispersive channels the LMMSE-optimal pilot pattern is obtained by solving an A-optimal sensor selection problem via either convex relaxation followed by randomized rounding or greedy selection, each combined with local swap refinement, and the resulting patterns consistently yield lower mean-square estimation error than rectangular or diamond lattices on practical block sizes.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["A-optimal selection beats lattices for OFDM pilots","Greedy selection beats lattices in OFDM pilot design","Convex relaxation optimizes OFDM pilot patterns","Local swaps refine pilots over fixed lattices in OFDM"],"cache_read_input_tokens":64,"weakest_assumption_plain":"The proposed heuristics reliably approximate the true A-optimal patterns and the doubly dispersive channel model together with the LMMSE criterion accurately reflect real-world performance on the finite grids considered.","fun_headline_variants_meta":{"raw":{"variants":["A-optimal selection beats lattices for OFDM pilots","Greedy selection beats lattices in OFDM pilot design","Convex relaxation optimizes OFDM pilot patterns","Local swaps refine pilots over fixed lattices in OFDM"]},"model":"grok-4.3","cost_usd":0.010528,"raw_usage":{"total_tokens":4518,"prompt_tokens":560,"num_sources_used":0,"completion_tokens":53,"cost_in_usd_ticks":105278000,"prompt_tokens_details":{"text_tokens":560,"audio_tokens":0,"image_tokens":0,"cached_tokens":64},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":3905,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":560,"tokens_out":53,"duration_ms":36949,"temperature":1.0,"reasoning_tokens":3905,"cache_read_input_tokens":64,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-08T19:21:49.355702+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A direct measurement of channel estimation mean-square error on a high-mobility wireless link that shows the proposed patterns produce no lower error than a rectangular lattice of equal pilot density.","supporting_citations":[],"review_version":1}