{"id":"a916fb80-d178-4a16-8e6f-75b546e8f0c1","arxiv_id":"2411.08264","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A sensing-assisted beam tracking scheme with codebook-based beamwidth adaptation raises simulated THz average rate and lowers outage probability for high-mobility targets.","lead":"This paper proposes a THz beam tracking scheme where the base station uses periodic sensing data to predict a mobile target's path and then picks a beamwidth-matched precoder from a precomputed codebook. Simulations report higher average data rates and lower outage probabilities than event-triggered and conventional beam tracking for fast-moving targets.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central claim depends on the assumption of perfect motion prediction: the paper's §II.B declares prediction error negligible and §IV evaluates on the very URM trajectory used to design the beam, so the claimed gains under real sensing noise or maneuver are untested.","rationale":"The paper's central claim—that real-time sensing can be converted into beamwidth choices that keep THz links connected at high mobility—would have to survive realistic sensing errors and motion-model mismatch. The condition that would make it true is that the target's path inside each sensing period is known exactly or with negligible error. That condition is assumed in §II.B and then used to construct both the codebook and the simulation trajectory, so the reported gains are internally consistent under the assumption but do not test it. The reader's weakest_assumption identifies exactly this error-free URM premise; I agree that it is the most load-bearing point. The concrete test I propose is a sensitivity analysis that perturbs the sensed state and the motion model, which is the minimal experiment that would separate the sensing-assisted beamwidth idea from the idealized prediction assumption. Because the paper itself does not include such an experiment, and because the manuscript's own text flags the prediction-error simplification as an assumption rather than a derived bound, the appropriate verdict remains CONDITIONAL: the scheme is plausible and the derivations are coherent, but the headline result is not yet robust to the uncertainty that sensing-assisted tracking is meant to address.","tokens_in":11087,"tokens_out":4770,"duration_ms":56545,"concrete_test":"Monte Carlo sensitivity test: keep all §IV parameters but add a sensing error model—e.g., angular position error ~N(0,σ) with σ = 0.5°, 1°, 2°, and/or a lateral acceleration of 2 m/s² or 5 m/s² so the target deviates from URM inside each τ interval. Recompute average rate and outage probability over at least 100 random sensing realizations for the proposed, event-based [7], and conventional schemes at v = 20, 50, and 100 m/s. Report Rmin, α, and PSO iterations as well. If the proposed scheme's outage exceeds 10% at 100 m/s, or its rate advantage over event-based falls within the error bars, the central claim does not hold without the perfect-prediction assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing premise is that the predicted angular path coincides with the actual path. In §II.B the paper states 'the prediction error is considered negligible, especially when τ is small' and treats sensed locations as accurate. The §IV simulations then evaluate the proposed scheme on the same URM trajectory used to construct δ and θm, so the beam is always centered on the true path. The performance gain over baselines is essentially a wide beam sized by exact knowledge of where the target will be; if sensing has nonzero error or the target accelerates or turns, the predicted angular interval is systematically shifted, and an interval-width beam cannot compensate for a wrong center. This is especially acute at τ=165 ms and v=100 m/s, where the target moves roughly 16.5 m per interval, so even modest angular sensing errors displace the beam significantly. The paper provides no sensitivity analysis, no error bars, and no comparison under mismatched motion. Secondary reproducibility gaps (unreported Rmin, penalty coefficient α, PSO settings, outage definition) compound the issue, but the core untested assumption is the error-free prediction on which the central claim rests.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a sensing-assisted beam tracking scheme for THz downlink communications. The base station periodically senses the target state, predicts a uniform rectilinear motion (URM) path over the sensing interval τ, and selects a precoder from a precomputed codebook. The precoder is restricted to the parameterized family f(δ, ω) in Eq. (7), whose main-lobe width is matched to the predicted angular interval, and the scalar parameter ω is optimized by particle swarm optimization (PSO) to maximize the time-average rate while penalizing violations of a minimum-rate constraint. A symmetry theorem is proved to halve the PSO search range. Simulations compare the proposed scheme with event-based beam tracking [7] and conventional tracking for target velocities from 10 m/s to 100 m/s, reporting higher average achievable rate and lower outage probability, with outage below 10% at 100 m/s.","tokens_in":11387,"tokens_out":6094,"duration_ms":63932,"significance":"If the reported gains survive when sensing is noisy and the target does not follow exactly the predicted URM trajectory, the scheme is a useful contribution to THz beam tracking for high-mobility scenarios. The closed-form beam-pattern integral in Eq. (11), the reduction of the precoder design to a single variable, and Theorem 1 are coherent, and the simulation study covers a reasonable range of velocities and transmit powers. The comparison includes a deliberate attempt to equalize sensing overhead between the proposed scheme and the event-based baseline. The main weaknesses are that the central evaluation relies on the assumptions of error-free sensing and exact URM, that several optimization parameters are unreported, and that no robustness or statistical variability analysis is provided. These issues are addressable within the scope of the paper, but they are load-bearing for the stated claims.","major_comments":[{"comment":"The central claim in Fig. 3(b) that the proposed scheme keeps outage below 10% even at 100 m/s rests entirely on the assumption that the sensed initial state is exact and that the target follows the same URM trajectory used to compute δ and θ_m in Eqs. (9)-(10). The paper states in Section II.B that sensed locations are \"considered accurate\" and that prediction error is \"considered negligible,\" and the simulations feed the design with exactly the trajectory on which performance is evaluated. At τ = 165 ms and v = 100 m/s the target moves 16.5 m per interval, so the beam center is very sensitive to angular sensing errors. Please add a sensitivity study with (i) noisy sensed position and velocity, (ii) acceleration or turning maneuvers, and (iii) prediction-horizon mismatch, and report whether the gains over [7] survive. This is not circular derivation, but it is a load-bearing robustness check for the stated claims.","section":"Section II.B and Section IV"},{"comment":"Several parameters that directly control the reported results are not disclosed: the minimum-rate threshold R_min, the penalty coefficient α in Eq. (16), the PSO swarm size, iteration count, and termination criteria, and the outage definition (presumably R < R_min, but this is never stated). Without these values the reader cannot reproduce Fig. 3 or verify that constraint C1 is actually satisfied rather than merely penalized. Please report all values and define outage probability precisely.","section":"Section III, Eq. (16), and Section IV"},{"comment":"No error bars, confidence intervals, or number of Monte Carlo runs are given. This matters because the event-based baseline has a random-walk component and the proposed scheme uses stochastic PSO; a single trajectory may not represent typical behavior, and the fluctuations in the event-based curve in Fig. 3(a) suggest run-to-run variability that should be quantified. Please state the number of independent runs and show variance or confidence bands.","section":"Section IV, Figs. 3-5"},{"comment":"The fairness adjustment τ = 3.3 × slot duration is derived from the average number of elapsed slots between beam realignments in [7], but the computation is not described in enough detail: is the average taken over the same velocity grid, over which random realizations, and is the same τ used for all three schemes at every velocity? Please clarify; otherwise the comparison in Fig. 3 may incorporate an uncontrolled dependence on the baseline's outage process.","section":"Section IV, Fig. 3"}],"minor_comments":[{"comment":"The first sentence of the scheme description says \"We proposed a three-step sensing-assisted beam tracking scheme\"; this should be \"We propose\" to match the present-tense description of the paper's contribution.","section":"Section II.B"},{"comment":"The symbol sinϕ_m is used without definition, and the expression \"sinϕ_m - ϕ(S)\" mixes two angle notations; please define the quantity consistently with ϕ(S) used in Eq. (3).","section":"Equation (19)"},{"comment":"The legend contains the typo \"sensing-asssited\"; Section IV also contains \"date rate\" for \"data rate.\"","section":"Fig. 3(b)"},{"comment":"The double summation is typeset with the index ranges in an unclear order (the limits over n and m are reversed relative to the summation order in the text); please fix the typesetting for readability.","section":"Equation (18)"},{"comment":"The terms \"optimal precoder\" and \"optimal beam\" should be qualified, since the optimization is over the restricted parameterized family in Eq. (7) and uses the penalty surrogate in Eq. (16), not a global solution of Q1.","section":"Abstract and Section III"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within scope for a communications journal and the core idea is reasonable, but the current evaluation idealizes sensing and motion to the point that the central robustness claim is untested. The requested sensitivity analysis, parameter disclosure, and statistical reporting are feasible within the scope of the paper, so I recommend major revision rather than rejection. No concerns about attribution or novelty were identified."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nShort version: this is a legitimate systems paper, not a breakthrough. The new piece is the joint use of periodic sensing to predict the target's angular path and then a parameterized precoder, taken from [7], whose beamwidth and shaping offset are optimized by PSO. Theorem 1 halves the search space via symmetry, and the proof is sound. The simulation setup is concrete—128-antenna ULA, 220 GHz, molecular absorption included—and the comparison to event-based [7] and conventional tracking is a reasonable effort, even if the baseline timing is calibrated to the baseline's own realignment rate.\n\nWhat it does well: the problem is clearly motivated, the optimization reformulation from Q1 to Q3 is sensible, and the paper does not oversell the novelty. It explicitly says the sensing information is accurate and that URM is assumed; the limitation is stated in Section II.B, not hidden. The rate and outage plots show consistent gains that are plausible given a perfectly known trajectory.\n\nThe soft spots are real but not fatal. The load-bearing premise is error-free prediction and no maneuvers. Since the simulation evaluates on exactly the URM path used to design the beam, the results are best-case. There is no sensitivity analysis for sensing noise or acceleration mismatch, and at 165 ms and 100 m/s the target moves 16.5 m per interval, so even a small angular error would shift the beam center. Also missing: the penalty coefficient α, PSO hyperparameters, the outage threshold Rmin, and error bars, which makes independent reproduction harder. These are omissions, not contradictions. The paper's claim is conditional on its own assumptions, and the authors state that condition.\n\nI agree with the reader's CONDITIONAL verdict. The concern about perfect prediction is valid, but it is an acknowledged idealization typical of first-cut systems papers, not a hidden circular flaw. The math holds, the extension is incremental but useful, and the citation pattern is fair—[7] is properly credited for the precoder form.\n\nWho this is for: researchers working on THz beam tracking or ISAC-assisted communications who want a concrete baseline for sensing-assisted beamwidth adaptation. It is citable. For peer review, I would send it out. A referee should ask for a robustness study under sensing error and a few missing parameters, but the core idea is sound and the presentation is honest.\n\nRecommendation: engage with it, but treat the numerical gains as upper bounds under idealized sensing.","headline":"A coherent, practical extension of beamwidth-adaptive THz tracking that leans on an unexamined perfect-prediction assumption, but deserves referee time.","tokens_in":11862,"tokens_out":1443,"would_cite":true,"duration_ms":17365,"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":"The paper shows that a sensing-assisted scheme that adapts beamwidth in real time to the target's predicted path achieves higher rates and outage below 10% at velocities up to 100 m/s.","keywords":["Terahertz communications","beam tracking","beamwidth adaptation","sensing-assisted communications","integrated sensing and communication","particle swarm optimization","high mobility","outage probability"],"falsifier":"Run the proposed scheme with a target that accelerates or changes direction within a sensing interval (or add Gaussian noise to the sensed position and velocity), and check whether the average rate still beats the event-based baseline and whether outage stays below 10% at 100 m/s; a controlled experiment with real radar or camera sensing feeding the same codebook would settle it.","tokens_in":10897,"feed_emoji":"📡","tokens_out":7678,"duration_ms":69991,"temperature":0.7,"pith_summary":"The paper proposes a THz communication scheme in which the base station periodically senses the target, predicts its path under a uniform rectilinear motion model, and then selects a precoder from a pre-calculated codebook whose beamwidth matches the predicted angular range. The precoder is parameterized by two scalars—coverage half-width δ and a beam-shaping offset ω—and is optimized once per sensing period to maximize the average achievable rate, with a penalty for dropping below a minimum instantaneous rate. The paper claims this real-time beamwidth adaptation yields higher average rates and lower outage probability than both event-based and conventional fixed-beam tracking, keeping outage below 10% even at target velocities of 100 m/s. If correct, the result means that ISAC-style sensing information can be converted directly into beamwidth choices that keep THz links stable in high-mobility scenarios such as vehicle-to-everything and drone communications.","feed_headline":"Sensing widens beams to keep THz links alive at 100 m/s","feed_subtitle":"Real-time motion prediction lets base stations trade beamwidth for coverage, holding outage below ten percent.","key_machinery":"The central object is the parameterized precoder f(δ,ω) = (β/(2δ)) ∫_{-δ}^{δ} u(p,ω) $e^{{jωp}}$ dp, a weighted linear combination of array response vectors over the predicted angular range, which concentrates the beam inside that range while letting δ (half-width) and ω (shape offset) control its coverage and flatness. Its n-th entry collapses to a sample function, f_n = β $e^{{-j(n-1)πθ_m}}$ Sa(δ(ω-(n-1)π)), making the beam shape a closed-form function of two scalars. The paper uses this parametrization to reformulate the constrained rate-maximization problem into a single-variable unconstrained one (with a penalty for the minimum-rate constraint), solved numerically by PSO to pre-build a codebook indexed by angular intervals. Theorem 1 proves the beamforming gain is axially symmetric in ω about (N_t-1)π/2, which halves the search space.","core_discovery":"With accurate periodic sensing and the assumption of uniform rectilinear motion, the authors show that constraining the beam to a shape that is flat over the predicted angular range—implemented by integrating array response vectors over that range with an offset ω—reduces the precoder design to a single-variable optimization solvable by particle swarm optimization. The resulting precoder, with n-th component f_n = β $e^{{-j(n-1)π θ_m}}$ Sa(δ(ω-(n-1)π)), automatically widens for fast targets and narrows for slow ones. Simulations with a 128-antenna array at 220 GHz show the scheme's average rate exceeds event-based and conventional tracking across velocities from 10 to 100 m/s, and its outage probability stays below 10% at the top speed. The authors position the contribution as a way to overlay motion awareness on beam tracking without changing the underlying sensing hardware.","pith_inferences":["If the target deviates from the assumed straight-line constant-speed motion—turning, braking, or being sensed with noise—our assessment is that the beam could center on the wrong path; a natural extension is a feedback loop that uses measured rate drops to hedge the predicted range wider.","The fairness setup sets τ to 3.3 times the event-based time slot to equalize overhead, implicitly assuming sensing costs the same as pilot-based realignment; in our view, a cost model that counts sensing energy and latency separately would sharpen the comparison.","Theorem 1's symmetry is specific to the ULA geometry; our guess is that a similar single-parameter search could be derived for uniform planar arrays by treating elevation and azimuth coverage half-widths separately, but that remains to be shown.","The paper assumes perfect sensing; a concrete robustness test we would run is to inject Gaussian errors into s0 and v0 and measure how quickly the outage probability climbs past 10%."],"forward_implications":["In a working system, the base station would execute only a codebook lookup per sensing period, moving the expensive PSO optimization offline and keeping switching latency within τ.","The approach extends beyond uniform rectilinear motion: swapping the motion model (e.g., uniformly accelerated motion) changes only the kinetic formula, so the same codebook machinery applies to accelerating or curving targets.","Because wider beams carry less gain, the scheme is a direct trade-off engine: it consciously sacrifices signal strength to avoid outages for fast targets, which is exactly what high-mobility THz links need.","The reported sub-10% outage at 100 m/s is a benchmark that ISAC-enabled THz systems can aim for when specifications call for reliable links to vehicles or UAVs."],"supporting_citations":[{"why":"Supplies the event-based beam tracking baseline with dynamic beamwidth adaptation that the proposed scheme must outperform, and the precoder formulation the paper extends.","marker":"[7]"},{"why":"Provides the line-of-sight THz channel model with free-space path loss and molecular absorption used in the rate calculations.","marker":"[4]"},{"why":"Establishes the integrated sensing and communication framework that motivates the assumption that the base station has access to periodic sensing information.","marker":"[11]"},{"why":"The particle swarm optimization algorithm used to solve the single-variable precoder optimization and build the codebook.","marker":"[19]"},{"why":"Supplies the closed-form model for THz joint radar-communication systems that underpins the claim that achievable rate depends on instantaneous target velocity.","marker":"[10]"},{"why":"Shows how prior information from channel estimation can track fast-varying THz channels by predicting target motion, the direct inspiration for using prior sensing for motion prediction.","marker":"[12]"},{"why":"Justifies neglecting NLoS and multi-path components at THz bands, which is why the array response and rate model the paper uses are valid.","marker":"[17]"}],"fun_headline_variants":["Real-time beamwidth adaptation keeps THz links stable at highway speeds","Sensing-driven beams adapt width to track 100 m/s targets in THz","Motion-predicting beams cut outage below 10% at 100 m/s","Particle-swarm-optimized precoders widen beams for fast THz users","Adaptive beamwidth codebook keeps THz links alive at 100 m/s"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The scheme's performance claims rest on the assumption that the target moves exactly as predicted from the sensed state—straight-line constant-speed motion with perfect sensing—so that prediction error is negligible; if targets accelerate, turn, or are sensed with noise, the beam may be aimed at the wrong place and the reported rate and outage gains could shrink or vanish.","fun_headline_variants_meta":{"raw":{"variants":["Real-time beamwidth adaptation keeps THz links stable at highway speeds","Sensing-driven beams adapt width to track 100 m/s targets in THz","Motion-predicting beams cut outage below 10% at 100 m/s","Particle-swarm-optimized precoders widen beams for fast THz users","Adaptive beamwidth codebook keeps THz links alive at 100 m/s"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00093,"raw_usage":{"total_tokens":3972,"prompt_tokens":926,"completion_tokens":3046,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":542,"completion_tokens_details":{"reasoning_tokens":2943}},"tokens_in":542,"tokens_out":3046,"duration_ms":20885,"temperature":1.0,"reasoning_tokens":2943,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T21:47:25.547748+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the proposed scheme with a target that accelerates or changes direction within a sensing interval (or add Gaussian noise to the sensed position and velocity), and check whether the average rate still beats the event-based baseline and whether outage stays below 10% at 100 m/s; a controlled experiment with real radar or camera sensing feeding the same codebook would settle it.","supporting_citations":[{"cited_title":"Event-based beam tracking with dynamic beamwidth adaptation in terahertz (thz) communications,","cited_arxiv_id":null,"evidence_quote":"Supplies the event-based beam tracking baseline with dynamic beamwidth adaptation that the proposed scheme must outperform, and the precoder formulation the paper extends."},{"cited_title":"A line-of-sight channel model for the 100–450 gigahertz frequency band,","cited_arxiv_id":null,"evidence_quote":"Provides the line-of-sight THz channel model with free-space path loss and molecular absorption used in the rate calculations."},{"cited_title":"Integrated sensing and communications: Toward dual-functional wire- less networks for 6g and beyond,","cited_arxiv_id":null,"evidence_quote":"Establishes the integrated sensing and communication framework that motivates the assumption that the base station has access to periodic sensing information."},{"cited_title":"Particle swarm optimization,","cited_arxiv_id":null,"evidence_quote":"The particle swarm optimization algorithm used to solve the single-variable precoder optimization and build the codebook."},{"cited_title":"Closed-form model for performance analysis of thz joint radar-communication systems,","cited_arxiv_id":null,"evidence_quote":"Supplies the closed-form model for THz joint radar-communication systems that underpins the claim that achievable rate depends on instantaneous target velocity."},{"cited_title":"Fast channel tracking for terahertz beamspace massive mimo systems,","cited_arxiv_id":null,"evidence_quote":"Shows how prior information from channel estimation can track fast-varying THz channels by predicting target motion, the direct inspiration for using prior sensing for motion prediction."},{"cited_title":"Multi-ray channel modeling and wideband characterization for wireless communications in the terahertz band,","cited_arxiv_id":null,"evidence_quote":"Justifies neglecting NLoS and multi-path components at THz bands, which is why the array response and rate model the paper uses are valid."}],"review_version":1}