{"id":"c2935295-f731-48ae-947c-67db0d7260d4","arxiv_id":"2504.18855","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":2.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A field-aware survey showing that near-field and cross-field propagation adds distance-based beam resolution, which changes scanning, estimation, beamforming, and tracking design.","lead":"This survey reviews beam management techniques for millimeter-wave and terahertz massive MIMO systems across far-field, near-field, and cross-field propagation. It argues that near-field and cross-field beams resolve users in distance as well as angle, and that field-aware design changes how beams are scanned, estimated, formed, and tracked.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified: the distance-resolution claim is supported by spherical-wave physics and by the paper's own derivations and cited literature; the HSPM parameter-sharing limitation is real but does not undermine the central survey claim.","rationale":"The reader's weakest-assumption analysis points to the HSPM parameter-sharing structure in Sec. II-A3. I agree that this is a genuine limitation: the paper itself notes that the HSPM parameters still need measurement characterization in [25], so the compact cross-field representation and the subarray-based estimation methods built on it are not yet fully validated. However, I do not see this as load-bearing for the central claim of the survey. The central claim, that NF and CF channels add distance-domain resolution relative to FF channels, is grounded in the spherical-wave channel model itself and is directly corroborated by the array-response phase expressions in (15)-(18) and (21)-(24), as well as by independent works cited throughout the paper. Even if the HSPM's equal-amplitude and common-reflector assumptions fail in some real THz channels, a cross-field channel would still exhibit distance-dependent phase differences between widely spaced subarrays whenever the user is in the near field of the overall array. The remaining issues noted by the reader, such as absent reproducibility artifacts for the figures and the absence of a new algorithmic result, are appropriate editorial conditions for a survey but do not change the correctness of the surveyed synthesis. I therefore recommend keeping the reader's CONDITIONAL verdict unchanged rather than moving it, and I do not identify an additional concern that would justify a stricter verdict.","tokens_in":43964,"tokens_out":9313,"duration_ms":106451,"concrete_test":"Reproduce Fig. 5 from the parameters stated in Sec. II-B (0.3 THz, LoS, N=1024, d=32λ, 16-subarray partition) and extend the comparison to half-wavelength element spacing; if the reported 12 dB HSPM advantage over PWM at 40 m is not reproduced, or if the HSPM advantage disappears at λ/2 spacing, the quantitative support for the HSPM accuracy claim would weaken.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I read the central claim as the qualitative assertion that NF and CF channels carry distance-dependent phase structure that FF planar-wave models lack, and that this structure matters for beam-management design. That claim is independently supported by the exact spherical-wave model in Sec. II-A1, by the beam-pattern expressions (18) and (24), and by the cited beam-focusing literature, so it does not rest on the HSPM parameter-sharing assumption in Sec. II-A3. The HSPM assumption of shared reflectors and equal path-gain amplitudes is explicitly flagged by the authors as awaiting measurement characterization in [25]; if real THz channels violate it, the compact parameter count and subarray-parameter-estimation methods built on HSPM would need revision, but the existence of distance resolution in NF/CF channels, which is the load-bearing part of the survey's conclusion, would survive because it follows from spherical-wave geometry. One nuance worth noting is that the survey itself reviews FF-beam-based NF acquisition [68] and field-independent beamforming [121], so the strong reading that all BM 'must' abandon FF methods is not the paper's actual position; the stated conclusion is that distance resolution is a key factor, which is consistent with the surveyed evidence. I therefore do not find a load-bearing flaw in the central argument.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript is a survey of beam management (BM) for millimeter-wave and terahertz massive MIMO, organized around propagation-field regimes. It introduces the spherical-wave (SWM), planar-wave (PWM), and hybrid spherical/planar-wave (HSPM) channel models, discusses their parameter counts and approximation errors, derives the corresponding beam patterns, and argues that near-field (NF) and cross-field (CF) beams provide resolution in both angle and distance, while far-field (FF) beams resolve angle only. The remainder of the paper reviews beam scanning, CSI estimation, beamforming, and beam tracking, classifying each by methodology and comparing how the techniques operate in FF, NF, and CF settings. The concluding claim is that the additional distance-domain resolution of NF/CF channels is a key factor that differentiates BM design across regimes.","tokens_in":44183,"tokens_out":15533,"duration_ms":156555,"significance":"The survey's central claim is well grounded. The exact SWM model in Section II-A1 and the beam-pattern expressions (18) and (24) show explicitly that NF/CF steering depends on distance as well as angle, and this is consistent with the cited near-field beam-focusing literature. The HSPM parameter-sharing assumption in Section II-A3 is acknowledged by the authors as awaiting measurement characterization in [25], and the main conclusion does not rest on that assumption. The taxonomy is useful, the comparison tables (Tables II–V) are a valuable reference, and the paper is unusually thorough in covering side-information-assisted and ML-based methods across all four BM functions. I found no internally inconsistent derivation that would undermine the paper's central qualitative message.","major_comments":[],"minor_comments":[{"comment":"The numerical values reported for the transition quantities (e.g., πd²LtsLrs/(λD11) ≈ 6.1×10⁻³ for 16 subarrays and πd²LtLr/(λD11) = 0.707 for Nt=Nr=1024 at 40 m) cannot be reproduced from the stated setting d=32λ under the definitions in Eq. (12); please reconcile the antenna-spacing parameter with the quoted numbers and report the full array and subarray geometry used in the simulations.","section":"Section II-B, Fig. 5"},{"comment":"The phase expression ∆Φ_i in Eq. (16) appears to have a unit or notation inconsistency with the extra factor d in e^{−j2πd/λ ∆Φ_i}; please clarify whether Γ and D are normalized by the antenna spacing or whether the factor d should be omitted.","section":"Section II-C, Eqs. (15)–(17)"},{"comment":"The claim that SWM and PWM are special cases of the HSPM at Kt=Nt, Kr=Nr and Kt=Kr=1 is correct in channel-matrix form, but the HSPM parameter count Np(1+5KrKt) does not reduce to the SWM parameter count 2NpNtNr in the former case; please clarify that the parameter-count comparison is intended for intermediate subarray partitions.","section":"Section II-A3"},{"comment":"The error analysis is performed only for the line-of-sight path; the statement that the extension to NLoS paths follows a similar procedure should be accompanied by a brief explanation of how reflector-dependent angles enter the error, or should be explicitly marked as an open validation issue.","section":"Section II-B"},{"comment":"The phrase 'building the basic through analyzing' should be rewritten as 'building the basics by analyzing'.","section":"Abstract and Section I-C"},{"comment":"The label 'Beam sacnning' should read 'Beam scanning'.","section":"Fig. 2"},{"comment":"The phrase 'production of the codeword and the array response vector' should be 'product of the codeword and the array response vector'.","section":"Section III, first paragraph"},{"comment":"The phrase 'while strike achieving high spectral efficiency' should read 'while still achieving high spectral efficiency'.","section":"Section V-A"},{"comment":"The word 'catheterized' should be 'categorized'.","section":"Section V-C"},{"comment":"References [75] and [77] are the same Dovelos et al. paper, and references [161] and [162] appear to be the same conference paper; please merge the duplicates and renumber the citations accordingly.","section":"References"}],"recommendation":"minor_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a well-organized survey that earns its keep as a reference map for field-aware beam management. The central claim—that near-field and cross-field channels add distance-domain resolution and that this matters for beam-management design—is supported by spherical-wave geometry and by the paper's own beam-pattern analysis; it does not rest on the HSPM parameter-sharing assumption. There are no new technical results, but that is not a flaw in a survey.\n\nWhat is actually new is the field-aware taxonomy and the comparison tables across scanning, CSI estimation, beamforming, and tracking. The channel-model and beam-pattern presentation is clear and in a unified notation. The paper is also honest about the HSPM: it explicitly says the shared-reflector and equal-amplitude assumptions await measurement characterization in [25]. Self-citations are present, but mostly for the HSPM and cross-field results where the authors are the right references.\n\nSoft spots, in proportion: the numerical figures lack reproducibility artifacts, and some setups (e.g., d=32λ in Fig. 5) are atypical and not fully parameterized, so those specific comparisons are hard to trust as given. The error analysis is line-of-sight only, and the HSPM-based illustrations inherit the untested sharing assumption. If real THz channels violate that structure, the HSPM's compact parameterization and the subarray-estimation methods built on it would need revision—but the distance-resolution conclusion, which is load-bearing for the survey's thesis, survives. The survey itself reviews far-field-beam-based near-field acquisition [68] and field-independent beamforming [121], so it does not overclaim that all beam management must abandon far-field methods.\n\nWho it is for: someone entering near-field/cross-field beam management, or anyone wanting one place to compare the methodological landscape. The math is sound where it matters, the citation pattern is appropriate, and the data limitations are clearly acknowledged. It deserves a serious referee; for a survey journal this is a normal conditional. Recommendation: engage with it.","headline":"A useful, honest survey of field-aware beam management; no new results, but the taxonomy holds and the central distance-resolution claim is sound.","tokens_in":44692,"tokens_out":2993,"would_cite":true,"duration_ms":30186,"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":"Near-field and cross-field channels give beam management a distance axis that far-field beams lack, and this survey shows how scanning, CSI estimation, beamforming, and tracking must be redesigned to exploit it.","keywords":["near-field communications","cross-field communications","beam management","terahertz MIMO","millimeter-wave MIMO","hybrid spherical and planar wave model","beam scanning","beam tracking"],"falsifier":"A channel measurement campaign at 0.3 THz with an ultra-large array (for example, 1024 elements) spanning distances from 5 to 40 meters could falsify the central claim: if the measured HSPM approximation error relative to the full spherical-wave channel does not stay small when each subarray lies within its own Rayleigh distance, or if subarray path gains and reflector sets are not shared, then the distance-domain beam management rationale loses its foundation.","tokens_in":1603,"feed_emoji":"📡","tokens_out":4541,"duration_ms":71141,"temperature":0.7,"pith_summary":"This survey argues that the decisive difference between far-field (FF) and near-field/cross-field (NF/CF) beam management is distance-domain resolution. In the FF regime, a beam is indexed by angle alone; in NF and CF regimes, the beam pattern also focuses in range, so scanning codebooks, channel estimation, beamforming, and tracking must treat angle and distance jointly. The paper develops this through channel-model analysis, introducing the hybrid spherical- and planar-wave model (HSPM) as a compact cross-field representation, and reviews beam management techniques across all four tasks. If the central claim is right, 6G millimeter-wave and terahertz systems cannot simply reuse far-field beam training procedures.","feed_headline":"Near-field links add a distance axis to beam management","feed_subtitle":"MmWave and terahertz MIMO beams that resolve range as well as angle change scanning, estimation, and tracking.","key_machinery":"The central object is the distance-dependent phase of the array response, captured in the cross-field setting by the hybrid spherical- and planar-wave model (HSPM). HSPM models spherical-wave propagation between subarrays and planar-wave propagation within each subarray, giving a parameter count of $N_p(1+5K_rK_t)$ instead of the quadratic count of the full spherical-wave model. The paper uses this model to show that the FF approximation error grows with $\\pi d^2 L_t L_r/(\\lambda D_{11})$, and that HSPM stays accurate when the communication distance exceeds the Rayleigh distance of each subarray. The HSPM-based beam pattern then depends on both angle and distance, which is what makes near-field and cross-field beam management different from far-field beam management.","core_discovery":"The central claim is that NF and CF channels provide additional resolution in the distance domain, and that this distance resolution is the key factor driving beam management design differences across propagation regimes. Whereas FF beams align along an angular direction, NF and CF beams focus at a specific position, meaning angle and distance are jointly resolved. This distinction propagates through every stage of beam management: codebooks must become polar or distance-aware, channel estimation must exploit spherical-domain sparsity, beamforming can separate users that share an angle but sit at different distances, and tracking must represent user position and velocity rather than just direction.","pith_inferences":["Editorial: if the HSPM is validated by measurement, it could become a standard low-complexity cross-field channel model for terahertz system design, a step the paper points toward but does not itself propose.","Editorial: the distance-resolution view suggests a testable benchmark: compare exhaustive, hierarchical, and ML-based scanning over a continuous range of distances crossing the Rayleigh boundary, rather than evaluating isolated FF and NF cases.","Editorial: co-directional users separated in range could be served simultaneously with focusing beams, turning distance resolution into a multiplexing resource in dense indoor terahertz hotspots."],"forward_implications":["Beam scanning codebooks for NF and CF links must cover both angle and distance, using polar-domain or hierarchical designs rather than angular-only far-field codebooks.","Channel estimation in NF and CF must exploit spherical-domain or polar-domain sparsity; far-field angular-sparsity assumptions degrade as the Rayleigh distance grows.","Beamforming with NF focusing beams can serve users at the same angle but different distances, which far-field steering beams cannot do without sacrificing signal power to one user.","Beam tracking in NF and CF should represent the state with user position and velocity (curvature of arrival) rather than only angles and path gain.","Extending far-field techniques to near-field and cross-field scenarios is possible but increases complexity, motivating unified, field-agnostic, low-overhead beam management strategies."],"supporting_citations":[{"why":"Provides the spherical-wave and planar-wave channel models that define the FF and NF propagation baselines.","marker":"[20]"},{"why":"Introduces the hybrid spherical- and planar-wave model (HSPM) that the survey adopts as the cross-field channel representation.","marker":"[22]"},{"why":"Channel measurement and modeling study the paper flags as needed to characterize the HSPM parameters.","marker":"[25]"},{"why":"Defines the Rayleigh distance that separates FF and NF regimes and underlies the distance-resolution argument.","marker":"[13]"},{"why":"Supplies the polar-domain near-field codebook used in scanning and channel estimation.","marker":"[37]"},{"why":"Shows far-field beam training can be redeployed for cross-field beam alignment, a key CF beam estimation result.","marker":"[68]"},{"why":"Provides a compressive-sensing-based cross-field channel estimation framework using a subarray codebook for the HSPM.","marker":"[86]"},{"why":"Establishes distance-domain orthogonality of near-field focusing beams, supporting NF beamforming and user separation.","marker":"[132]"}],"fun_headline_variants":["Distance becomes a beam axis in mmWave MIMO","Near-field beams resolve distance, not just angle","Beam management gets a distance dimension","Cross-field MIMO: beams focus, not just point","Distance resolution reshapes beam management"],"cache_read_input_tokens":46848,"weakest_assumption_plain":"The argument leans on the assumption that, in the hybrid spherical- and planar-wave model, different subarrays share the same reflectors and path-gain amplitudes and differ only in phase and the angles of departure and arrival; if real terahertz cross-field channels do not have this sharing structure, the cross-field channel representation and the beam-management conclusions built on it weaken.","fun_headline_variants_meta":{"raw":{"variants":["Distance becomes a beam axis in mmWave MIMO","Near-field beams resolve distance, not just angle","Beam management gets a distance dimension","Cross-field MIMO: beams focus, not just point","Distance resolution reshapes beam management"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000883,"raw_usage":{"total_tokens":3811,"prompt_tokens":938,"completion_tokens":2873,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":554,"completion_tokens_details":{"reasoning_tokens":2804}},"tokens_in":554,"tokens_out":2873,"duration_ms":21423,"temperature":1.0,"reasoning_tokens":2804,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T10:07:44.382464+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A channel measurement campaign at 0.3 THz with an ultra-large array (for example, 1024 elements) spanning distances from 5 to 40 meters could falsify the central claim: if the measured HSPM approximation error relative to the full spherical-wave channel does not stay small when each subarray lies within its own Rayleigh distance, or if subarray path gains and reflector sets are not shared, then the distance-domain beam management rationale loses its foundation.","supporting_citations":[{"cited_title":"Multiple Access for Near-Field Communications: SDMA or LDMA?","cited_arxiv_id":null,"evidence_quote":"Establishes distance-domain orthogonality of near-field focusing beams, supporting NF beamforming and user separation."}],"review_version":1}