{"id":"2157de4e-ffec-4156-9c9a-e61a708c7be6","arxiv_id":"2608.11651","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A single CMOS chip uses frequency-controlled leaky-wave metasurface antennas to steer terahertz beams, enabling simultaneous multi-user communication and centimeter-level localization for multi-agent systems.","lead":"This paper reports a 208 to 258 gigahertz transceiver chip, built in standard silicon, that steers radio beams in different directions just by changing the signal frequency and can talk to several robots at once. It combines communication and radar-like localization on a single 1.5 by 4.9 millimeter chip, a step toward small wireless systems for future indoor robot teams.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The simulated-only DGS isolation for TX-RX cancellation is the most load-bearing unverified assumption; a breakout S21 measurement would settle it.","rationale":"Good-faith reading: the paper is a proof-of-concept ISAC transceiver with frequency-space mapping, and most claims have measured support. The most important claim that is not backed by a direct measurement is the DGS isolation. The reader's weakest assumption identifies exactly this gap, and I agree. I weighed the alternative concerns: the 210-Gbps aggregate is a disclosed sum of per-channel rates and is not presented as a concurrent rate in the data section, so it is a reporting issue rather than the central physical claim; the AoA method is limited to parallel apertures but this is explicitly stated (Fig. 5a, 'theta_TX = theta_RX = theta') and the localization experiment provides an independent distance check. The DGS concern is load-bearing because simultaneous TX/RX is the unique enabler of the multi-agent SFDMA demonstration, and no direct measurement of the coupling path exists. The one-to-three experiment and the adjacent-channel SNR sweep provide indirect evidence, which is why I would not reject or unverify the paper; I would keep the reader's CONDITIONAL verdict, with the condition being a direct DGS isolation measurement or a more stressful adjacent-channel concurrent-link test.","tokens_in":33726,"tokens_out":12256,"duration_ms":131733,"concrete_test":"Fabricate a dedicated DGS breakout test structure containing the same two HLM apertures with the same 70-um edge-to-edge spacing and the same defected-ground geometry, but with wafer-probe pads at the internal HLM input ports, and measure S21 from the TX aperture input to the RX aperture input over 205-265 GHz. Compare the measured isolation with the simulated 38-61 dB values in Supplementary Fig. S1-8 and with the LNA blocking margin at each frequency in Fig. 2i. Agreement within about 5 dB would resolve the concern; a systematic shortfall would require revising the simultaneous TX/RX and concurrent multi-link claims.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing assumption is the DGS-based TX-to-RX isolation that enables simultaneous TX/RX operation (Supplementary Section 1.3, Fig. 2h-i, Supplementary Fig. S1-8/S1-10). The claimed 38-61 dB isolation is obtained entirely from full-wave simulation; the paper explicitly states that on-chip aperture-to-aperture coupling cannot be directly measured at the aperture ports. The system-level validation is an adjacent-channel SNR sweep, which tests one desired RX channel at a time under TX excitation in neighboring channels. The one-to-three concurrent SFDMA experiment uses well-separated channels (CH9/CH19/CH35), so it does not stress the lowest isolation margins or the worst-case frequency proximity. If the fabricated DGS cancellation deviates from simulation due to process variation, metal fill, or quartz-superstrate assembly, the simulated LNA blocking margin of 12.82 dB (Fig. 2i) could be eroded or reversed, directly degrading the central concurrent multi-link and simultaneous TX/RX claims. This is a verification gap rather than an observed error, and the disclosed adjacent-channel SNR stability (0.8-1.2 dB variation) is genuine supporting evidence, but it is not a direct measurement of the isolation that the argument depends on.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript reports a 208-258 GHz 65-nm CMOS transceiver chip that monolithically integrates TX and RX front ends with two heterogeneous leaky-wave metasurface (HLM) apertures. The HLM maps carrier frequency to radiation angle via the N=-2 Floquet harmonic, producing a measured 75-degree scan with only four meta-atoms. The paper demonstrates 38 frequency-spatial channels, an aggregate capacity of 210 Gbps obtained by summing per-channel rates, a one-to-three concurrent SFDMA link experiment at 6 Gbps, a prototype-similarity AoA estimator with a mean error of 0.672 degrees, and 2D localization with a mean error of 1.49 cm. A defected-ground-structure (DGS) decoupling structure is used to suppress TX-to-RX coupling and enable simultaneous transmit/receive operation, with the isolation characterized by full-wave simulation and indirect system-level SNR measurements.","tokens_in":33855,"tokens_out":7921,"duration_ms":85879,"significance":"The work is a significant systems-level advance for integrated THz ISAC: it is the only chip in the comparison to combine on-chip beamforming, single-chip TX/RX integration, and integrated THz PA/LNA front ends, and the dispersion model provides a genuine prediction that is checked against full-wave simulation and independent radiation-pattern measurements. Credit is due to the calibrated over-the-air characterization, the explicit link-budget equations, the cross-validation of the AoA estimator including an unseen trajectory, and the transparent disclosure that the 210 Gbps figure is a sum of per-channel rates. The main risks are the simulation-only DGS isolation and the labeling of the 210 Gbps value as an aggregate; both are verification and claim-precision issues rather than observed errors in the underlying data.","major_comments":[{"comment":"The 210-Gbps aggregate is not a concurrent aggregate. The main text states that it is obtained by summing the individually demonstrated per-channel rates, and the only concurrent multi-link experiment (one-to-three) delivers 6 Gbps. Table 1 places '210 Gbps' in the Aggregate data rate column next to a checkmark for concurrent multi-link operation, which conflates the two metrics. To make the claim precise, the paper should label the 210 Gbps value as the sum of per-channel peak rates, report the 6 Gbps concurrent result as the demonstrated aggregate, and either remove the checkmark or add a footnote clarifying the distinction.","section":"SFDMA-enabled multi-agent THz communication; Table 1"},{"comment":"The DGS-based TX-RX isolation is quoted as 38-61 dB with a 12.82-dB LNA blocking margin, but this isolation is obtained entirely from full-wave simulation; the paper explicitly states that the on-chip aperture-to-aperture coupling cannot be measured directly. The adjacent-channel SNR sweep (Supplementary Fig. S1-10) and the one-to-three experiment provide real but indirect system-level evidence, and the CH9 sweep is a useful worst-case test. However, the one-to-three experiment uses channels separated by 10 and 16 channel spacings, so it does not stress the worst-case frequency proximity, and the simulated blocking margin is load-bearing for the concurrent TX/RX claim. I recommend adding a direct DGS-isolation test structure or, failing that, presenting the one-to-three experiment as the primary evidence for simultaneous TX/RX and explicitly downgrading the simulation-only isolation figures accordingly.","section":"Supplementary Section 1.3; Fig. 2h-i"}],"minor_comments":[{"comment":"The angle convention in Eq. (6) should be stated explicitly; the scan range is quoted as -33 degrees to +42 degrees, and a clear definition of theta with respect to broadside would remove ambiguity for readers.","section":"Eq. (6); Fig. 3c"},{"comment":"The AoA estimator is described as 'neural-network-assisted,' but the model is a prototype-similarity weighted interpolation with 101 trainable parameters; the authors should either explain the neural-network terminology or use a more precise term such as 'learned prototype regression.'","section":"Supplementary Section 4.3"},{"comment":"The data and code are available only upon request; for reproducibility, the authors should deposit the AoA dataset, the training code, and the measured channel data in a permanent repository.","section":"Data and code availability"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a strong systems-level demonstration that fits the journal's applied-physics scope. The two major comments concern the precision of the aggregate-rate claim and the verification gap for the simulated DGS isolation; both are addressable within a revision. I have no concerns about citation practices or novelty disclosure, and I would be comfortable with acceptance after the authors address these points."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a real engineering advance—a single 65-nm CMOS die that puts TX/RX front ends and frequency-scanned leaky-wave metasurface apertures together, and backs it with careful over-the-air measurements. The 210-Gbps aggregate is not a concurrent throughput, though the authors do disclose that; treat it as a sum of per-channel bests. The bigger soft spot is that the TX-RX isolation enabling simultaneous operation rests on simulation, not a direct measurement.\n\nWhat's new: the HLM aperture with four meta-atoms (CIW/PPW/meandered MSL) is a clever way to get 75° scan at 208–258 GHz without element-level control, and the Floquet dispersion prediction matches measured beam angles—genuine full-wave-to-experiment agreement. The amplifier-last TX / amplifier-first RX with the TEON gain-boosted PA/LNA is solid circuit work; the PA numbers compare well. The sensing side is more than a gesture: the PDF-CDF distance-normalized features make physical sense, the AoA regressor is a simple two-parameter prototype-similarity model with proper five-fold CV, and the 2D trajectory uses unseen data. The one-to-three SFDMA demo (two uplinks, one downlink at 2 Gbps each) is the first concurrent multi-link result of this kind. The paper is also admirably transparent about what was not done: no concurrent 38-channel operation, boundary samples excluded from AoA validation.\n\nSoft spots, in order of size. First, the Table 1 entry lists 210 Gbps aggregate with a checkmark for concurrent multi-link, which alongside the abstract could mislead a casual reader into thinking the chip does 210 Gbps concurrently. The text is clear, but the table should be annotated or the headline changed to 'sum of per-channel rates.' This is a reporting issue. Second, the DGS-based isolation (38–61 dB) is entirely from CST; the authors state the on-chip aperture coupling can't be measured directly. The adjacent-channel SNR sweep (0.8–1.2 dB variation) is decent indirect evidence, and the three-link demo works, but a breakout S21 measurement on a future iteration would remove the main residual doubt. The concern is real but not a reason to doubt the demonstrated system behavior. Third, the isolation margin is smallest at CH9, and the concurrency demo uses CH9/CH19/CH35—so the worst-case frequency proximity isn't stressed in the multi-link test. The per-channel adjacent tests cover that, but not simultaneously.\n\nVerdict: this deserves peer review. The claims are mostly supported, the limitations are disclosed, and the integration is novel. Send it to a knowledgeable referee, and ask for a breakout isolation measurement or a clear justification of why it's infeasible, plus a revision of the aggregate-rate presentation. I'd cite it if I work in THz ISAC.","headline":"A genuine single-die THz ISAC integration advance—frequency-scanned leaky-wave apertures with concurrent multi-link and localization—but the 210-Gbps aggregate is a sum of per-channel rates and the TX-RX isolation rests on simulation; still deserves peer review.","tokens_in":34520,"tokens_out":3095,"would_cite":true,"duration_ms":32544,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A single 65-nm CMOS die maps 208–258 GHz to radiation angle through four-meta-atom leaky-wave metasurfaces, enabling 38 spatial-frequency channels, concurrent links, sub-degree AoA, and centimetre-level 2D localization.","keywords":["terahertz ISAC","leaky-wave metasurface","frequency-to-space mapping","spatial-frequency division multiple access","CMOS transceiver","beam scanning","angle-of-arrival estimation","2D localization"],"falsifier":"Fabricate a test die that routes the TX and RX HLM feed ports to probe pads and measure the S21 between them, with and without the slotted ground structure; the claimed simultaneous-TX/RX operation requires measured isolation near the simulated 38–61 dB across 208–258 GHz.","tokens_in":33438,"feed_emoji":"📡","tokens_out":9387,"duration_ms":93231,"temperature":0.7,"pith_summary":"This paper reports a fully integrated terahertz transceiver chip that aims to make multi-agent wireless communication and mutual spatial awareness share one tiny CMOS die. It claims that four 'meta-atoms' of a heterogeneous leaky-wave metasurface can map each carrier frequency to a unique radiation angle, replacing phased-array beamforming networks with frequency dispersion. If true, a swarm of small robots could talk to several peers simultaneously over different frequency-steered beams and locate each other to within centimetres without extra antenna hardware. The chip is demonstrated over the air at 208–258 GHz with 75 degrees of beam scan and 38 frequency-spatial channels, with individually measured per-channel rates summing to 210 Gbps.","feed_headline":"Tiny chip maps 208-258 GHz to 75 degrees of beam scan","feed_subtitle":"One tiny chip turns frequency into beam direction for multi-agent THz links and centimetre-level mutual localization.","key_machinery":"The load-bearing object is the heterogeneous leaky-wave metasurface (HLM), a periodic aperture of four meta-atoms, each formed by chip-integrated waveguide, parallel-plate waveguide, and S-shaped microstrip sections. Its periodicity acts as a spatial mixer: the guided wave's propagation constant $eta$ is combined with the lattice wavevector $K$ to form Floquet harmonics $eta_N = \\beta + NK$, and the $N=-2$ harmonic falls inside the light cone and radiates. This spatial-domain mixing is cascaded with the frequency-domain mixing of the mixer/PA/LNA chain, so that selected IF tones become selected free-space beams. A defected-ground structure (DGS) between the two apertures adds an artificial coupling path designed to cancel the intrinsic TX-to-RX coupling, letting transmitter and receiver run simultaneously.","core_discovery":"The paper's central claim is that one 1.5 × 4.9 mm² die in 65-nm CMOS can monolithically combine active terahertz transmit and receive front ends with frequency-scanned apertures, so that the carrier frequency, not a phase-shifter network, selects the beam direction. The heterogeneous leaky-wave metasurface (HLM) alternates chip-integrated waveguide, parallel-plate waveguide, and S-shaped microstrip sections in each of four meta-atoms; its periodicity produces Floquet harmonics, and the $N=-2$ harmonic radiates. Measurements show a 75° scan as frequency sweeps 208–258 GHz, with 38 frequency-spatial channels; the paper reports three simultaneous links, a mean angle-of-arrival error of 0.672°, and a mean 2D localization error of 1.49 cm. The aggregate 210 Gbps is the sum of individually demonstrated per-channel rates.","pith_inferences":["A natural extension is to put a second periodic direction into the same HLM, which should let one aperture resolve both azimuth and elevation; the paper lists this as future work but does not demonstrate it.","Because the AoA feature is normalized by total received power, the same estimator should keep its accuracy at other link distances and between the 1° training grid angles; that can be tested without new hardware.","The link-budget equations imply that raising transmit power by PA power-combining and adding a lens or larger aperture would extend the 82-cm link toward room-scale coverage; that extrapolation goes beyond the measured data."],"forward_implications":["Each link in a multi-agent network can be assigned its own frequency-spatial channel, and the one-to-three demonstration shows concurrent uplinks and a downlink at 2 Gbps each.","Direction finding and communication use the same aperture: a 50-tone probe creates an angle-dependent received spectrum, and the neural estimator recovers angles with mean absolute error below 0.7°.","Aperture complexity no longer grows with the number of beams: four meta-atoms give the full 75° scan, so adding users adds tones rather than phase shifters or antenna elements.","Using the AoA estimate to select the carrier aligns both ends of a link, improving image transmission over random carrier selection.","The same die supports both a long-distance mode (1 Gbps BPSK over 82 cm) and a high-rate mode (25 Gbps 32-QAM over 8 cm), letting channel bandwidth be traded against distance."],"supporting_citations":[{"why":"D-band CMOS joint radar-communication transceiver providing the single-link ISAC baseline this work extends.","marker":"[12]"},{"why":"220-GHz portable phased-array ISAC system that reaches 64 Gbps but only for a single user; comparison point for concurrent multi-link operation.","marker":"[13]"},{"why":"Leaky-wave metasurface and antenna concepts that provide the frequency-to-direction mapping principle later embodied in the HLM.","marker":"[42-44]"},{"why":"Earlier CMOS on-chip leaky-wave antenna with 220–500 GHz bandwidth and 128° scanning; demonstrates that leaky-wave apertures can be fabricated in the same process.","marker":"[45]"},{"why":"On-chip topological leaky-wave antenna for full-space THz connectivity, showing wireless links with a leaky-wave aperture but with external THz coupling.","marker":"[46]"},{"why":"THz prism CMOS system using leaky-wave antennas and transceivers for simultaneous localization; the direct predecessor for the AoA and localization demonstrations.","marker":"[48]"},{"why":"Frequency-division multiplexing in the THz range using a leaky-wave antenna; supports the SFDMA channel concept.","marker":"[49]"},{"why":"Single-shot link discovery for THz networks; motivates frequency-addressed link establishment.","marker":"[50]"}],"fun_headline_variants":["Tiny chip steers THz beams by frequency, not phase shifters","One chip: 75° beam scan from 208-258 GHz","CMOS chip maps frequency to space for THz links and sensing","Frequency-driven beam steering on a 1.5×4.9 mm THz chip","THz transceiver chip uses frequency to aim beams and locate agents"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper assumes that the slotted ground structure between the two on-chip antennas cancels transmitter-to-receiver coupling as the simulations say it does; the supplement states that the coupling cannot be measured directly at the aperture ports on the fabricated die, so if the physical cancellation falls short of the simulated 38–61 dB, simultaneous transmit and receive would degrade.","fun_headline_variants_meta":{"raw":{"variants":["Tiny chip steers THz beams by frequency, not phase shifters","One chip: 75° beam scan from 208-258 GHz","CMOS chip maps frequency to space for THz links and sensing","Frequency-driven beam steering on a 1.5×4.9 mm THz chip","THz transceiver chip uses frequency to aim beams and locate agents"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000627,"raw_usage":{"total_tokens":2904,"prompt_tokens":951,"completion_tokens":1953,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":567,"completion_tokens_details":{"reasoning_tokens":1854}},"tokens_in":567,"tokens_out":1953,"duration_ms":14834,"temperature":1.0,"reasoning_tokens":1854,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:32:35.741384+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Fabricate a test die that routes the TX and RX HLM feed ports to probe pads and measure the S21 between them, with and without the slotted ground structure; the claimed simultaneous-TX/RX operation requires measured isolation near the simulated 38–61 dB across 208–258 GHz.","supporting_citations":[],"review_version":1}