{"id":"a7d9a2d9-e8e2-4873-aa8d-4428ba8dc8de","arxiv_id":"2505.16807","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":3,"one_line_summary":"A chirp radar waveform that embeds data in delay, Doppler, and amplitude, plus a chirp-division multiple access scheme, is proposed for low-rate automotive sensing and communication.","lead":"This paper proposes a way for cars to use their existing chirp radar signals to send low-rate control messages to other vehicles while continuing to sense the road. The authors describe a resource-sharing plan and two modulation schemes, and they provide simulation results and code for the Doppler-division version.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The decision-directed tracking loop that separates data-induced shifts from physical motion is never analyzed or simulated; without it the claimed simultaneous ISAC is unsupported, so the REJECT verdict stands.","rationale":"The reader's weakest assumption is the decision-directed tracking loop, and I agree this is the single most load-bearing concern. The entire data demodulation concept in Section III-B relies on the receiver's ability to distinguish data-induced delay/Doppler perturbations from physical motion using an EKF prediction. The paper offers only a qualitative inequality (constellation spacing > sensing parameter variations) and explicitly lists the feedback between demodulation and tracking as an open problem in Section IV-B. Without a closed-loop simulation or an analytic error-propagation bound, there is no evidence that the scheme works at the claimed -35 dB SNR or survives target maneuvers. This concern directly undermines the abstract's claim that the scheme enables simultaneous sensing and communication. I also note the Chirp-DMA \"interference-free\" assertion in Section III-A lacks a derivation and multi-pair simulation; however, the tracking loop is more fundamental because it is required for even a single transceiver pair. The REJECT verdict is appropriate, and a revision could become CONDITIONAL by adding a closed-loop tracking simulation, a multi-pair interference simulation, and a baseline comparison with FRaC or OTFS-based ISAC.","tokens_in":11679,"tokens_out":9083,"duration_ms":79268,"concrete_test":"Simulate a closed-loop decision-directed receiver over many frames: the AT transmits QPSK-modulated delay/Doppler symbols each CPI; the PT runs an EKF predicting the next delay/Doppler, demodulates the residual to the nearest constellation point, removes the data, and updates the track; compute BER and track-loss probability as functions of target acceleration (0, 2, 5 m/s^2) and SNR (-35, -30, -25 dB). Compare against a genie-aided receiver that knows the true position. If closed-loop BER diverges from genie-aided BER (e.g., error floor or >3 dB degradation), the decision-directed assumption is falsified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section III-B states: \"Communication data is first extracted during the tracking process through the mixture of sensing parameters and communication data\" and the method works \"as long as the minimum distance between the designed constellation symbols is larger than the variations in sensing parameters.\" This is a decision-directed loop: the receiver predicts where the transmitter should appear in the delay-Doppler map, demodulates the residual as data, then uses the data-removed measurement to update the track. No error-propagation analysis, no closed-loop simulation, and no quantitative threshold for \"variations in sensing parameters\" are given. In a realistic V2V scenario, the relative position and velocity between AT and PT change continuously; the modulation shifts the measured delay/Doppler by discrete resolution cells. If the tracking prediction error exceeds half a constellation spacing, demodulation fails. A single symbol error then corrupts the measurement used for the track update, so errors can cascade into track loss. The paper's claimed operation at -35 dB makes this particularly fragile, and Section IV-B even admits \"larger errors in parameter estimation result in larger tracking errors, which in turn adversely affect the demodulation performance.\" The simulations in Fig. 5 do not specify whether the reference position was genie-aided or closed-loop, and no multi-frame simulation is presented. Therefore, the central claim that radar sensing and communication can be performed simultaneously with only FFT/CFAR/MUSIC is not established.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a chirp-based integrated sensing and communication (ISAC) framework for autonomous vehicles. It introduces Chirp-DMA, a cognitive-resource-allocation scheme that identifies idle time-frequency resources using dedicated short-duration chirps and claims interference-free operation for multiple transceiver pairs as long as the chirps are separated by a certain delay. Within each allocated resource block, the paper proposes two delay-Doppler domain modulation schemes: a TDM-based scheme with per-PRI delay and amplitude modulation plus per-CPI Doppler modulation, and a DDM-based scheme with CPI-level modulation of delay, Doppler, and complex amplitude. Both schemes embed data in the same delay-Doppler parameters used for sensing, to be extracted during tracking with FFT, CFAR, MUSIC, and EKF. The DDM scheme is evaluated by simulation through BER, hitrate, and estimation-error CDFs at SNR values down to -35 dB, with data rates of a few kbps. The paper also discusses open problems including synchronization, malicious interference, multi-vehicle networking, and extension to embodied agents, and provides a GitHub link for simulation code.","tokens_in":11908,"tokens_out":4171,"duration_ms":37260,"significance":"If the central claims were substantiated, the paper would offer a practically attractive ISAC design: a low-complexity extension of existing FMCW radar signal processing that simultaneously provides sensing and low-rate V2V communication without dedicated communication hardware or spectrum. The use of familiar radar algorithms (FFT, CFAR, MUSIC, EKF) and the provision of simulation code are strengths, and the paper is honest about the low data rates and about several idealizing assumptions. However, the key claims are not supported by the evidence presented. The interference-free property of Chirp-DMA is asserted rather than derived or simulated; the tracking-based demodulation principle is a decision-directed loop whose stability is neither analyzed nor simulated; and the data-rate formulas assume a full delay modulation range that is not reconciled with the physical target position. The simulation section omits essential setup details and does not demonstrate the multi-frame closed-loop operation that the demodulation concept requires.","major_comments":[{"comment":"The claim that 'interference-free ISAC can be achieved as long as there is a certain delay among different chirps' is not substantiated. No signal model for the multiple-access scenario is given, no analysis of the cross-correlation between a victim receiver's locally generated chirp and an interfering chirp with a different start time or chirp rate is provided, and no simulation of multiple transmitter-receiver pairs is shown. The derived capacity T_c/(2T_u) relies on an idealized timing grid (Fig. 3) whose robustness to non-ideal synchronization, range-dependent path loss, and near-far effects is asserted rather than demonstrated. Because Chirp-DMA is the stated mechanism behind the abstract's 'without interference' claim, this is a load-bearing gap.","section":"Section III-A"},{"comment":"The demodulation principle is a decision-directed tracking loop: data bits are read from the received delay-Doppler mixture using sensing predictions, and the data-removed measurement is then used to update the track. The paper states that 'Communication data is first extracted during the tracking process through the mixture of sensing parameters and communication data' and that this works 'as long as the minimum distance between the designed constellation symbols is larger than the variations in sensing parameters.' No error-propagation analysis is given, no quantitative threshold for 'variations in sensing parameters' is derived, and no closed-loop multi-frame simulation of the tracking-demodulation loop is presented. Section IV-B explicitly concedes that 'larger errors in parameter estimation result in larger tracking errors, which in turn adversely affect the demodulation performance.' Without a stability or error-propagation analysis and without end-to-end simulation of this loop, the claimed simultaneous sensing and communication at SNR down to -35 dB is unsupported.","section":"Section III-B"},{"comment":"The data-rate formulas assume that the full N_s/2 delay range is available for modulation, independent of the target's true position. In Fig. 4, the target has both an 'actual detected position' and a 'should-be' position, which implies that some range bins are occupied by the physical target and cannot be used for data symbols without ambiguity or false-target risk. The paper does not specify how the transmitter avoids the target's true range bin, how the receiver knows which bins are allowable, or how the modulation range is renegotiated when the target delay changes between frames. The claimed data rates therefore overstate the achievable delay-domain capacity.","section":"Section III-B"},{"comment":"The simulation section reports BER, hitrate, and CDF curves but omits essential setup details: the number and parameters of targets, radar cross-section, multipath and channel model, noise and SNR reference point, the number of Monte Carlo runs, and whether the reference position used for demodulation is genie-aided or obtained from the tracking loop. Fig. 5(a) is described in the caption as 'data demodulation performance' but the text refers to BER; the curves are not distinguished by marker style. Without these details, and without a signal model or equation for the DDM waveform, the provided code link cannot be independently checked, and the central claim of reliable operation down to -35 dB cannot be assessed.","section":"Section III-C"}],"minor_comments":[{"comment":"There are typos: 'enabe' in the abstract and 'estimiation' and 'cumulative distributive function' in Section III-C should be 'estimation' and 'cumulative distribution function', respectively.","section":"Abstract and Section I"},{"comment":"The notation N_s, f_s, and N_c is used in the data-rate formulas but N_s is not explicitly defined before its first use; a formal list of symbols would improve readability.","section":"Section III-B"},{"comment":"The statement that the TDM BER curve will shift 6 dB to the right compared with the DDM scheme is not supported by any simulation or derivation shown in the paper.","section":"Section III-C"},{"comment":"The figure caption does not identify line styles or symbols for the different configurations, making it hard to map the curves to the legend entries; a table of configurations and data rates would help.","section":"Fig. 5"},{"comment":"The comment that the worst-case BER is 1 rather than 0.5 because 'if the detection fails, no demodulation takes place' is not fully explained in a conventional BER framework and should be clarified or justified.","section":"Section III-C"}],"recommendation":"reject","confidential_remarks":"The manuscript is more a systems-concept paper than a fully validated technical contribution; its quantitative claims (interference-free multi-user operation, -35 dB SNR operation, data rates) are not backed by derivations or complete simulations. The provided code link is a positive sign, but the paper would need a substantially rewritten validation section, including a signal model for the multiple-access scenario and an end-to-end simulation of the tracking-demodulation loop, before it could be considered publishable in a journal. The editor may also wish to consider whether the level of technical validation is appropriate for IEEE Network."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Hi [Name],\n\nYou should know two things about arXiv:2505.16807 before anything else: the core idea is genuinely practical—modulating data on delay and Doppler shifts in a standard chirp radar, read out with the same FFT/CFAR/MUSIC chain—and the paper never actually validates the one mechanism that makes it work. The receiver separates data from physical motion using a tracking prediction, but there's no error-propagation analysis and no closed-loop simulation. Section IV-B admits the coupling. That's a load-bearing gap.\n\nWhat's new: the Chirp-DMA resource-sensing layer, where dedicated short chirps sniff idle time-frequency slots, is a nice addition to the FRaC-style index-modulation idea. The paper is clearly written and does a good job explaining why chirp-based ISAC needs to be sensing-centric. The TDM/DDM distinction is sensible, and the complexity analysis is useful.\n\nWhere it falls short: the interference-free claim for Chirp-DMA is asserted, not shown—no derivation, no multi-pair simulation. The simulation section omits essential details: channel model, target model, noise statistics, Monte Carlo runs, and whether the tracking reference was genie-aided or estimated. The BER curves for delay, Doppler, and amplitude are reported as identical, which is suspicious and not explained. The data-rate formulas assume the full N_s/2 delay range is available regardless of target position, which can't hold near the maximum range. The code link is welcome, but without a commit hash it's hard to verify. These aren't cosmetic problems; they concern the central feasibility of the scheme.\n\nAll that said, the paper is not nonsense. The idea is plausible and the authors are honest about open problems, including synchronization and malicious interference. With a proper signal model, multi-pair interference simulation, and a genuine closed-loop tracking/demodulation simulation, this could become a solid contribution. As it stands, the claims outrun the evidence.\n\nFor a reading group, it's worth a maybe. I wouldn't cite it as a result, but I'd send it to a serious referee because the concept deserves scrutiny and the field needs low-complexity V2X ISAC ideas. The right outcome is likely major revision or reject, not desk reject.","headline":"A practical low-rate ISAC concept for chirp radar, but the decision-directed tracking loop that makes it work is never validated.","tokens_in":12507,"tokens_out":3497,"would_cite":false,"duration_ms":26983,"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":"This paper proposes Chirp-DMA, a chirp-division multiple access scheme that lets mmWave radar pairs simultaneously sense targets and exchange low-rate control data, using only standard radar processing.","keywords":["integrated sensing and communication","chirp-division multiple access","mmWave radar","delay-Doppler domain modulation","vehicle-to-everything","autonomous driving","MIMO radar","range-Doppler map"],"falsifier":"Run the DDM scheme in simulation with a tracked target whose true delay jitter is made comparable to the minimum constellation spacing, for example by adding a second reflection path whose strength is within a few dB of the direct path, and measure BER and tracking divergence across many frames. If the decision-directed demodulation loop cannot keep the error below the constellation spacing under this condition, the paper's central assumption fails.","tokens_in":11463,"feed_emoji":"📡","tokens_out":6410,"duration_ms":50494,"temperature":0.7,"pith_summary":"This paper sets out to show that an ordinary chirp-based mmWave radar can double as a low-rate communication link without changing its sensing hardware or its standard processing chain. The authors propose a chirp-division multiple access scheme in which each transceiver first listens with a short dedicated chirp to find idle time-frequency slots, then modulates data onto the delay, Doppler, and amplitude of its chirp echoes. They claim that as long as chirps from different users arrive with a minimum delay separation, multiple radar pairs can run integrated sensing and communication at the same time without mutual interference. The scheme is designed to use only FFT, CFAR, and MUSIC, and the DDM variant is simulated to work at signal-to-noise ratios down to about -35 dB. If true, this would give autonomous vehicles a low-cost way to exchange control messages with nearby cars while continuing to sense them.","feed_headline":"Chirp radar doubles as a low-rate data link","feed_subtitle":"Multiple mmWave radar pairs can sense targets and swap control data on the same chirps, without interference or extra hardware.","key_machinery":"The load-bearing object is the dedicated chirp, a locally generated chirp with the same chirp rate as the transmitted signal but shorter duration $T_u = T_c f_{cut}/B$, used to identify idle time-frequency resources by mixing incoming signals and detecting IF tones. The associated Chirp-DMA allocates tilted-bar resource blocks with a guard time $T_u$, so that $T_c/(2T_u)$ transceiver pairs can operate concurrently. Inside each block, the modulation mechanism is to displace the detected peak in the range-Doppler map relative to the position predicted by tracking: a quantized delay shift encodes data in range, a quantized Doppler shift encodes data in velocity, and a QAM or PSK symbol encodes data in complex amplitude. The receiver separates communication data from sensing parameters by relying on the fact that sensing parameters change continuously while data symbols are discrete, provided the minimum constellation spacing exceeds the sensing variation.","core_discovery":"The central claim is that the chirp itself can carry data: by shifting the position of a detected target peak in the range-Doppler map by a small, quantized amount in delay or Doppler, or by changing the complex amplitude of the chirp, a transmitter can embed a few bits per frame without changing the waveform's radar-friendly properties. Because sensing parameters evolve slowly and predictably, the paper argues that the receiver can first subtract the predicted delay and Doppler from the measured peak position, decode the residue as communication data, and then use the remaining parameters to continue tracking. Two modulation variants are given: a TDM-MIMO version that modulates delay and amplitude every pulse repetition interval and Doppler once per coherent processing interval, reaching hundreds of kilobits per second, and a DDM-MIMO version that modulates all three dimensions once per CPI, reaching only kilobits per second but working at lower SNR because all antennas transmit simultaneously. The proposed Chirp-DMA uses a dedicated chirp shorter than the regular chirp to probe which tilted time-frequency bars are occupied, and the authors state that interference-free ISAC is achievable as long as there is a certain delay among different chirps, with up to $T_c/(2T_u)$ transceiver pairs accommodated. The paper validates the DDM version by simulation, showing that data demodulation, target detection, and delay, velocity, and angle estimation all succeed at low SNR under the chosen parameter settings.","pith_inferences":["Beyond the paper, the low rates mean the chirp link is limited to control-plane messaging; high-volume sensor sharing would need a separate channel.","Beyond the paper, a realistic stress test is multipath with strong secondary reflections, because the tracking loop could misread a delay jump as data; the paper does not simulate this.","Beyond the paper, constellation spacing could be adapted to tracking confidence to trade rate for robustness, an option the paper leaves open."],"forward_implications":["With Chirp-DMA, multiple mmWave radar transceiver pairs can share the same time-frequency resources, and the number of simultaneously supported pairs is set by the ratio of the chirp duration to the dedicated-chirp duration.","The DDM-based scheme gives kbps-level rates, enough for vehicular control messages, while the TDM-based scheme raises the rate by roughly the number of chirps per frame, at a 6 dB SNR penalty.","The whole receiver chain is built from FFT, CFAR, and MUSIC, so the added communication function does not require new hardware or a separate radio.","Because bandwidth, PRI, and SNR are interchangeable, doubling bandwidth or PRI buys 3 dB of equivalent SNR, allowing a vehicle to trade range resolution or update rate for link margin.","The scheme operates at SNR values near -35 dB, suggesting lower transmit power and less interference to other spectrum users."],"supporting_citations":[{"why":"Supplies the standard FFT and CA-CFAR radar processing chain that the proposed scheme builds on.","marker":"[1]"},{"why":"Provides the MIMO radar orthogonal-waveform framework, including TDM and DDM, used to separate transmit antennas.","marker":"[3]"},{"why":"Identifies proactive radar interference mitigation and the idea that time offsets can separate chirps, which Chirp-DMA extends without a dedicated control channel.","marker":"[11]"},{"why":"Gives the hitrate performance metric and a prior joint radar-communication system via index modulation that the simulations compare against.","marker":"[14]"},{"why":"Supplies the MUSIC super-resolution angle estimation algorithm used in the receiver.","marker":"[15]"},{"why":"Reviews chirp parameter modulation schemes such as ASK, FSK, and PSK that the proposed delay-Doppler modulation extends from SISO to MIMO.","marker":"[9]"}],"fun_headline_variants":["Chirp radar embeds data in delay-Doppler shifts","Radar chirps carry extra bits for V2X without extra hardware","New chirp modulation merges sensing and communication in cars","Chirp-DMA: radar pairs sense and talk on same waveform","Low-rate data link rides on mmWave radar chirps"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole method rests on the receiver's ability to tell a data-induced shift from a motion-induced shift by trusting predictions from the tracking filter, with the paper assuming that the minimum distance between constellation symbols is larger than the variation in sensing parameters; no analysis is given for what happens when tracking errors, multipath, or a missed detection violate that inequality.","fun_headline_variants_meta":{"raw":{"variants":["Chirp radar embeds data in delay-Doppler shifts","Radar chirps carry extra bits for V2X without extra hardware","New chirp modulation merges sensing and communication in cars","Chirp-DMA: radar pairs sense and talk on same waveform","Low-rate data link rides on mmWave radar chirps"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000654,"raw_usage":{"total_tokens":3103,"prompt_tokens":1159,"completion_tokens":1944,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":775,"completion_tokens_details":{"reasoning_tokens":1855}},"tokens_in":775,"tokens_out":1944,"duration_ms":11520,"temperature":1.0,"reasoning_tokens":1855,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T14:54:55.312342+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the DDM scheme in simulation with a tracked target whose true delay jitter is made comparable to the minimum constellation spacing, for example by adding a second reflection path whose strength is within a few dB of the direct path, and measure BER and tracking divergence across many frames. If the decision-directed demodulation loop cannot keep the error below the constellation spacing under this condition, the paper's central assumption fails.","supporting_citations":[{"cited_title":"MIMO radar for advanced driver- assistance systems and autonomous driving: Advantages and challenges,","cited_arxiv_id":null,"evidence_quote":"Provides the MIMO radar orthogonal-waveform framework, including TDM and DDM, used to separate transmit antennas."},{"cited_title":"Radar interference mitigation for automated driving: Exploring proactive strategies,","cited_arxiv_id":null,"evidence_quote":"Identifies proactive radar interference mitigation and the idea that time offsets can separate chirps, which Chirp-DMA extends without a dedicated control channel."},{"cited_title":"FRaC: FMCW- based joint radar-communications system via index modulation,","cited_arxiv_id":null,"evidence_quote":"Gives the hitrate performance metric and a prior joint radar-communication system via index modulation that the simulations compare against."},{"cited_title":"Sensing user’s channel and location with terahertz extra-large reconfigurable intelligent surface under hybrid-field beam squint effect,","cited_arxiv_id":null,"evidence_quote":"Supplies the MUSIC super-resolution angle estimation algorithm used in the receiver."},{"cited_title":"Joint wireless communication and radar sensing systems–state of the art and future prospects,","cited_arxiv_id":null,"evidence_quote":"Reviews chirp parameter modulation schemes such as ASK, FSK, and PSK that the proposed delay-Doppler modulation extends from SISO to MIMO."}],"review_version":1}