{"id":"1bdfd786-83d8-4daf-bc6a-f6594e7f9084","arxiv_id":"2508.04627","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Closed-form Cramér-Rao bounds for near-field joint angle-and-distance sensing, plus beamforming designs, show that hybrid hardware and energy-efficiency gains degrade distance-estimation accuracy in ISAC.","lead":"A 6G antenna array that serves phones can double as a radar, and this paper works out the best beam patterns for doing both when targets sit in the near field. It matters because it quantifies a tradeoff: saving power and using cheaper hardware measurably worsens how well the array can judge a target's distance.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Supplied full text is not the paper under review (arXiv:2508.04624 [math.AC] in mojibake), so the closed-form CRB/BCRB derivations that carry the central claim are absent and unverifiable; the claim currently rests on an abstract.","rationale":"The reader's verdict is UNVERDICTED with confidence LOW, based on the abstract alone and the inability to read the full text. My stress-test confirms that the most load-bearing condition—the correctness of the closed-form CRB/BCRB derivations—cannot be checked from the supplied material. The attached full text has a different arXiv identifier and is mojibake, so the technical core is effectively missing. Per the reviewing rule, I treat this as an explicit missing-support flag rather than dismissing it as a pipeline artifact. The reader identified the near-field channel model as the weakest assumption; that is a plausible and related concern, but the more immediate problem is that not even the channel model or derivation is accessible. I therefore partially agree: the reader's identified assumption is one manifestation of the larger unverifiability. A concrete check is to obtain the real paper and independently rederive the CRB. Since my concern does not introduce new information beyond the reader's, the verdict should remain UNCHANGED—UNVERDICTED, pending access to a readable manuscript. No internal inconsistency can be established from the abstract; the concern is about missing evidence, not proven error.","tokens_in":18693,"tokens_out":4293,"duration_ms":52437,"concrete_test":"Fetch the actual PDF for arXiv:2508.04627 from arXiv (not the supplied text); locate the system model and the CRB derivation; independently recompute the Fisher information matrix for the point-target near-field model by symbolically differentiating the spherical-wavefront array response with respect to angle and distance, invert it, and compare the resulting closed-form CRB entries with the paper's equations at a representative parameter set. If the expressions do not match, the central claim is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"To support 'we first derive the closed-form CRB of joint angle-and-distance estimation ... and the Bayesian CRB of the target response matrix,' the paper must provide a correct Fisher information matrix, a valid inversion, and an optimization that respects hybrid analog/digital constraints. The attached body is undecodable mojibake and carries arXiv:2508.04624v1 [math.AC], which does not match arXiv:2508.04627 cs.IT. Thus none of the derivation, the system model, the convexification steps, or the simulation baselines can be inspected. The abstract, while coherent, is not evidence for algebraic correctness. If the derivations contain an error, all headline conclusions—near-field joint angle-distance feasibility, hybrid degradation of distance accuracy, EE-accuracy tradeoff—would not follow. This is a missing-support problem, not a claim about author intent. The reader's channel-model worry is one specific manifestation: we cannot check the spherical-wavefront response or its identifiability. The most load-bearing concern is therefore that the central claim's proof is absent from the submission.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The abstract describes an ISAC paper that derives closed-form Cramér–Rao bounds (CRB) for joint angle-and-distance estimation of a point target and a Bayesian CRB (BCRB) for an extended target, then minimizes these bounds through hybrid beamfocusing under energy-efficiency and QoS constraints, using penalty-based successive convex approximation and alternating optimization, with simulations reporting that near-field joint estimation is feasible, hybrid architectures degrade distance-estimation accuracy, and higher EE reduces target-estimation accuracy. However, the supplied full text is not a readable version of this paper: it is mojibake and carries the header 'arXiv:2508.04624v1 [math.AC] 6 Aug 2025', which does not match the claimed arXiv:2508.04627 (cs.IT). No system model, channel model, CRB/BCRB derivation, optimization algorithm, or simulation result can be inspected. The central technical claims are therefore unsupported by the submitted manuscript.","tokens_in":18886,"tokens_out":4687,"duration_ms":51194,"significance":"If the derivations and simulations were correct, the paper would provide a useful design framework: closed-form bounds for near-field joint angle-distance estimation, an optimization procedure for hybrid beamfocusing, and concrete comparative statements about hybrid-versus-digital and EE-versus-accuracy tradeoffs. The relative comparisons are a sensible way to mitigate the self-referential nature of bound-based design. However, because the body is unreadable and mislabeled, no derivation, algorithm, or simulation result can be credited. There are no machine-checked proofs or reproducible code in the submission. The potential significance is high, but the current manuscript does not permit verification of any claimed contribution.","major_comments":[{"comment":"The body is entirely undecodable and is labeled 'arXiv:2508.04624v1 [math.AC] 6 Aug 2025', not the claimed cs.IT paper. The abstract's central assertions—closed-form CRB/BCRB, joint estimation feasibility, hybrid degradation, EE-accuracy tradeoff—require derivations and simulations that are absent. This is a missing-support problem, not a scientific disagreement; I cannot inspect the Fisher information matrix, its inversion, or the optimization constraints.","section":"Full text (all pages)"},{"comment":"Because no equations are readable, the claims 'we first derive the closed-form CRB ... and the Bayesian CRB ...' and 'joint distance-and-angle estimation is feasible' are unsupported. Identifiability depends on the near-field array response and on the Fisher information matrix being nonsingular; neither can be checked. A complete manuscript should include the full channel model, the FIM, the inversion, and the conditions under which the FIM is nonsingular. As submitted, this is a load-bearing omission.","section":"Abstract / Full text (absent derivations)"},{"comment":"The optimization minimizes CRB/BCRB and the abstract evaluates outcomes using the same bounds. This self-referential loop is partially mitigated if the comparisons against fully-digital baselines and different EE operating points use independent evaluation, but the paper should state explicitly whether any result is based on Monte Carlo estimation or only on the bounds. The absence of this clarification is secondary to the missing text, but it is relevant to interpreting the headline claims.","section":"Design/evaluation (abstract)"}],"minor_comments":[{"comment":"The arXiv identifier printed in the body should be corrected to 2508.04627 [cs.IT]; the current header shows 2508.04624v1 [math.AC].","section":"Header/full text"},{"comment":"If resubmitted, the document encoding must be clean; the current body is unreadable, preventing any check of notation, equations, or simulation plots.","section":"Full text (encoding)"}],"recommendation":"reject","confidential_remarks":"This appears to be a wrong file upload or a severe encoding failure; the submission is not reviewable in its current state. I recommend rejection of this submission. If the correct manuscript is available, the editor may wish to invite a fresh submission, but the version submitted cannot support any positive editorial decision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The one thing you should know: the full text attached to this review is undecodable mojibake and carries the identifier arXiv:2508.04624v1 [math.AC], not arXiv:2508.04627 cs.IT. I could only evaluate the abstract. That is the biggest soft spot, and it is not minor.\n\nIf the abstract is accurate, the paper does something genuinely useful. It claims the first closed-form CRB for joint angle-and-distance estimation in near-field ISAC, the first Bayesian CRB for an extended-target response matrix, and an optimization framework that minimizes those bounds under hybrid beamforming, energy-efficiency, and QoS constraints. The three headline findings—joint estimation is feasible in the near field, hybrid architecture degrades distance accuracy, and higher EE costs sensing accuracy—are specific, testable, and coherent. They are the kind of results designers would actually use.\n\nBut the central evidence is missing. The closed-form CRB/BCRB derivations, the Fisher information matrix, the convexification, the simulation baselines—none of it is inspectable in this copy. That matters because CRB algebra is exactly where subtle errors live, and the feasibility claim reduces to nonsingularity of the FIM under the spherical-wavefront model. The channel model is the weakest premise, as the reader noted, but I cannot even check whether the paper states it properly. The circularity concern (minimizing the CRB and then evaluating with the same CRB) is real but mild, since the headline results are relative comparisons against fully-digital baselines. There is also no visible code or error bars, but I cannot assign much weight to that without seeing the simulations.\n\nCitation pattern and related-work engagement are not assessable from this submission.\n\nMy take: this is a promising paper that is not reviewable in its current form. A correct PDF for arXiv:2508.04627 should be requested from the authors. If the derivations check out under referee scrutiny, it deserves publication as a solid ISAC contribution. Right now, citing it or bringing it to reading group would be based on an abstract alone. I would desk-reject this corrupted submission, but explicitly invite resubmission of the readable version—do not let the format failure kill the work.","headline":"The supplied manuscript is unreadable and mismatched to the arXiv ID, so the paper's load-bearing CRB/BCRB derivations are unverifiable; the abstract is promising but cannot carry the claims.","tokens_in":19497,"tokens_out":2451,"would_cite":false,"duration_ms":30358,"reading_group":"maybe","serious_thinker":"unclear","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper derives closed-form bounds for near-field ISAC sensing accuracy and optimizes hybrid beamfocusing to minimize them under energy and QoS constraints.","keywords":["integrated sensing and communication","near-field","hybrid beamforming","beamfocusing","Cramér-Rao bound","Bayesian CRB","energy efficiency","massive MIMO"],"falsifier":"Measure angle and distance estimation error with a real antenna array at several signal-to-noise ratios and compare the empirical mean-squared error against the predicted CRB: if the empirical error falls below the CRB, or if the model's Fisher information matrix is singular while a practical estimator still resolves range, the modeling premise is wrong. A simpler version: sweep target range toward the Fraunhofer distance and check whether the distance CRB diverges as predicted.","tokens_in":18499,"feed_emoji":"📡","tokens_out":3323,"duration_ms":41658,"temperature":0.7,"pith_summary":"This paper tries to establish that near-field integrated sensing and communication systems, where large antenna arrays make targets sit in the near-field region, can estimate a target's angle and distance together, and that the achievable accuracy can be predicted in closed form. It derives the Cramér-Rao bound (CRB) for joint angle-and-distance estimation from a point target and the Bayesian CRB (BCRB) for the target response matrix of an extended target, then optimizes transmit beamfocusing to minimize these bounds while maintaining system energy efficiency and communication quality of service. The paper argues that joint distance-and-angle estimation is feasible in the near field, but hybrid analog-digital architectures degrade distance accuracy relative to fully digital beamformers, and raising system energy efficiency degrades target estimation accuracy. If the paper is right, designers get a formula-level handle on what near-field sensing can and cannot do before building the hardware.","feed_headline":"Near-field ISAC can estimate distance and angle together","feed_subtitle":"Closed-form bounds tie sensing accuracy to energy efficiency and hybrid hardware choices.","key_machinery":"The central object is the near-field array response: a spherical-wavefront vector whose phase at each antenna element depends on both the target angle and its distance from that element. This distance-dependent phase curvature is what makes range identifiable, and it is what lets the Fisher information matrix for angle and distance be nonsingular. The paper builds closed-form CRB/BCRB expressions from this response and then treats the bound as the optimization objective for transmit beamfocusing, using penalty-based successive convex approximation for the fully digital design and alternating optimization to split the analog and digital beamformer design.","core_discovery":"The central claim is that in the near-field regime the received signal depends on both the target angle and its distance through the spherical wavefront across the array, and this joint dependence makes angle-and-distance estimation feasible in the sense that the Fisher information matrix is nonsingular. The paper derives a closed-form CRB for joint angle-and-distance estimation for a point target and a Bayesian CRB for the target response matrix for an extended target. It then uses these bounds as the objective for transmit beamfocusing design, solving the resulting nonconvex problems with a penalty-based successive convex approximation for the fully digital case and an alternating optimiza","pith_inferences":["Editorial inference: the same closed-form bounds could be inverted for system sizing—given a required sensing accuracy, one could solve for the minimum array aperture, bandwidth, or number of RF chains needed.","Editorial inference: since range identifiability comes from wavefront curvature, the distance CRB should diverge as the target moves toward the conventional far-field boundary; sweeping target range and plotting the distance CRB would be a direct test of the feasibility claim.","Editorial inference: the reported energy-efficiency versus accuracy tradeoff suggests a Pareto-frontier formulation that the paper does not fully characterize; mapping that frontier would let operators pick operating points between sensing precision and battery life."],"forward_implications":["The closed-form CRB and BCRB can serve as performance benchmarks for any near-field ISAC estimator and any candidate beamformer.","Under the assumed model, switching from fully digital to hybrid beamforming is predicted to preserve angle accuracy while measurably reducing distance accuracy, giving system designers a concrete cost for hardware savings.","Increasing system energy efficiency is predicted to directly raise the sensing error bound, making the energy-accuracy tradeoff an explicit design parameter rather than an afterthought.","The beamfocusing designs can hold communication quality-of-service above required thresholds while steering sensing accuracy to a desired level."],"supporting_citations":[],"fun_headline_variants":["Near-field ISAC: joint angle and distance estimation unlocked","Energy-efficient near-field ISAC trades accuracy for power","Hybrid beamfocusing degrades distance sensing in near-field ISAC","Closed-form bounds tie near-field ISAC sensing to energy and hardware","Near-field ISAC: distance and angle from one beam"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The received signal follows exactly the assumed spherical-wavefront array response, with both target angle and distance appearing in the phase at every antenna element; anything that breaks that phase model—mutual coupling, element phase errors, or targets outside the near-field regime—collapses the CRB derivation and the feasibility claim.","fun_headline_variants_meta":{"raw":{"variants":["Near-field ISAC: joint angle and distance estimation unlocked","Energy-efficient near-field ISAC trades accuracy for power","Hybrid beamfocusing degrades distance sensing in near-field ISAC","Closed-form bounds tie near-field ISAC sensing to energy and hardware","Near-field ISAC: distance and angle from one beam"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000812,"raw_usage":{"total_tokens":3445,"prompt_tokens":839,"completion_tokens":2606,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":583,"completion_tokens_details":{"reasoning_tokens":2521}},"tokens_in":583,"tokens_out":2606,"duration_ms":21066,"temperature":1.0,"reasoning_tokens":2521,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T23:52:11.205439+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure angle and distance estimation error with a real antenna array at several signal-to-noise ratios and compare the empirical mean-squared error against the predicted CRB: if the empirical error falls below the CRB, or if the model's Fisher information matrix is singular while a practical estimator still resolves range, the modeling premise is wrong. A simpler version: sweep target range toward the Fraunhofer distance and check whether the distance CRB diverges as predicted.","supporting_citations":[],"review_version":1}