{"id":"874323ec-6a79-4b75-9f80-96b2a687694d","arxiv_id":"2607.15930","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"VaporISAC uses one molecular vapor waveform for both data transmission and environmental sensing, and demonstrates the idea with a diffusion-advection simulation and reuse of an ethanol-based molecular communication testbed.","lead":"This paper proposes VaporISAC, a system that uses puffs of chemical vapor to carry data and simultaneously reveal information about the surrounding air, such as airflow or smoke. It is a concept-and-feasibility proposal for environments where radio signals fail, like rubble or smoke-filled tunnels.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Experimental sensing claim lacks ground truth and confound control; peak shifts cannot be uniquely attributed to airflow given MQ-3 response dynamics and unmeasured variables.","rationale":"The reader identified the idealized 1D transport model as the weakest assumption, which is valid. My stress-test sharpens this into an experimental validation gap: even within the controlled testbed, the sensing extraction is not isolated from confounds. The paper's own Section VII acknowledges sensor drift, nonlinearity, and lack of validation, which corroborates that the experimental demonstration does not yet establish the advertised environmental inference. This is a correctness-risk concern, but it does not invalidate the framework as a proposal; the analytical example is internally consistent and the concept remains plausible. Thus the appropriate verdict remains CONDITIONAL, with the condition being a controlled, ground-truthed experimental validation. Since the reader already set CONDITIONAL, no verdict change is needed (UNCHANGED). My partial agreement reflects that I emphasize the experimental confounds and sensor dynamics more than the 1D-vs-3D modeling gap, though both point to the same central weakness: the sensing claim is not yet demonstrated.","tokens_in":7463,"tokens_out":3409,"duration_ms":36239,"concrete_test":"Rerun the experimental proof-of-concept with independent ground truth: place a calibrated anemometer at the receiver and use a precision syringe pump for vapor release. (1) Vary airflow from 0.2 to 1.5 m/s at fixed release mass, estimate drift velocity from peak delay using the paper's method, and compare against the anemometer. (2) Vary release mass at fixed airflow to test whether amplitude-based attenuation estimates are confounded. (3) Characterize the MQ-3 step response and deconvolve sensor dynamics from the measured concentration. If estimated velocities track ground truth across >=3 airflow levels and are insensitive to release mass, the concern is resolved. Complementary check: generate waveforms from the 1D model with unknown v, D, A, and sensor time constant, then test whether a joint estimator uniquely recovers v without known transmitted bits.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that a single molecular waveform simultaneously carries data and enables environmental inference (airflow, turbulence, losses) without dedicated pilots. The analytical demonstration (Fig. 3) uses a known 1D diffusion–advection model with preset parameters, so it shows only self-consistency, not identifiability in real conditions. The experimental proof (Fig. 5) reuses testbed data from [14] with no ground-truth airflow measurement, no controlled release mass, and no calibration of the MQ-3 sensor's nonlinear response or time constant. Observed peaks arrive ~44% earlier and are ~66% stronger under 'drift-assisted' propagation, but these differences could equally stem from sensor memory, release amount, humidity, or baseline drift—especially since MQ-3 response times are on the order of seconds, comparable to the reported peak shifts. The paper's own Section VII lists 'sensor drift, nonlinear sensor response, and strong temperature dependence' as open challenges. Therefore the conclusion that environmental parameters have been 'demonstrated' to be inferred from communication waveforms is not supported by the presented evidence; the sensing inference rests on an unvalidated assumption rather than a controlled measurement.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript introduces VaporISAC, a framework for integrated sensing and communication (ISAC) based on macroscopic molecular communication. The central proposal is that a single chemical vapor pulse can serve simultaneously as a communication symbol and as an environmental probe, so that the received concentration waveform is jointly processed to recover data and to infer properties such as airflow velocity, turbulence, propagation losses, and chemical conditions. The paper motivates the approach by EM-ISAC limitations, describes a transmitter–channel–receiver architecture, presents a 1D diffusion–advection analytical example with OOK waveform '101101', re-uses a previously reported MQ-3 testbed measurement as an experimental illustration, discusses applications in robotics, industrial monitoring, and search-and-rescue, and ends with an open-challenges section.","tokens_in":7802,"tokens_out":4795,"duration_ms":56926,"significance":"The conceptual contribution is genuinely novel and timely: applying the ISAC principle to the chemical/molecular domain is an underexplored direction, and the paper provides a clear architectural picture, a useful comparison table, and an honest statement of open problems in Section VII. The analytical OOK example is transparent and uses standard diffusion–advection formulas with explicit parameters, which is a strength. However, the sensing inference is not yet validated: the analytical curves are generated from the same model used to interpret them, and the experimental part lacks ground-truth airflow and controlled conditions. If the claims are appropriately softened and the experimental evidence is reframed as illustrative rather than validating, the paper would make a worthwhile contribution to the molecular communication and ISAC communities.","major_comments":[{"comment":"The experimental sensing claim is not supported by the presented data. The reported estimate of 0.314 m/s for 'effective drift velocity' is not compared to any independent ground-truth airflow measurement, and the testbed was originally built for communication, not sensing. The observed changes in peak arrival time and amplitude could be caused by uncontrolled factors such as release volume, sensor response dynamics, humidity, or baseline drift, especially since MQ-3 response times are on the order of seconds. No repeated trials or error bars are reported. Given that the Conclusion states that environmental properties were 'demonstrated' from the same molecular waveform, this is a load-bearing overclaim. The authors should either provide controlled experiments with independent airflow measurement and multiple repetitions, or explicitly relabel Fig. 5 as a qualitative illustration.","section":"Section V, Fig. 5"},{"comment":"The analytical proof-of-concept is internally consistent but does not establish the identifiability of environmental parameters. The waveforms are generated from the same 1D diffusion–advection model used to interpret them, and the parameters v1, v2, D, A1, A2, and the emitted mass are all preset. In practice, peak arrival time depends on v, D, and release timing, while peak amplitude depends on v, D, emitted mass, and attenuation; the figure's claim that peak times 'reveal airflow velocity' and peak amplitudes 'indicate propagation losses' assumes that the other parameters are known. No estimation, sensitivity, or identifiability analysis is provided. Please add an explicit uncertainty/identifiability discussion, or formulate the claim more modestly as 'under known D and emission conditions, these waveform features carry information about v and losses.'","section":"Section V, Fig. 3"},{"comment":"There is a direct internal inconsistency about the status of the experimental evidence. Section V says 'To experimentally validate this concept,' and the Conclusion states that the paper 'demonstrated through analytical and experimental proof-of-concept demonstrations' that a single waveform can reveal airflow and losses. Yet Section VII, under 'Hardware, Safety, and Validation,' identifies 'the lack of experimental validation' as a major barrier to VaporISAC maturation. These statements cannot both stand. The manuscript should be revised so that the experimental part is described consistently—either as preliminary, uncontrolled data that motivated the framework, or as a full validation with ground truth.","section":"Sections V and VII"}],"minor_comments":[{"comment":"The caption and text use A1 and A2 but A1 is never defined. Please specify that A1 is the reference attenuation (1.0) and justify the choice of A2 = 0.55.","section":"Fig. 3"},{"comment":"The x-axis is 'Time [s]' and the y-axis is 'Concentration [a.u.]', but the y-axis extends to 3 with no explanation of the normalization. Also, 'approximately44%' is missing a space and the estimated velocity of 0.314 m/s is given without a confidence interval or measurement uncertainty.","section":"Fig. 5"},{"comment":"The entries for VaporISAC communication range (0.5–5 m) and data rate (bps) appear to be assumed rather than derived from theory or measurements. Please provide a source or a clarifying sentence that these are target orders of magnitude.","section":"Table I"},{"comment":"The paper does not describe the exact signal processing that would jointly recover bits and estimate environmental parameters beyond 'fixed threshold' and 'peak picking.' A more explicit description of the claimed joint decoder would strengthen the presentation.","section":"Section V"},{"comment":"Reference [12] is a two-page NANOCOM abstract; the sentence citing it says it demonstrates simultaneous binary decoding and source localization. Please verify that this source indeed supports that claim and briefly describe how the sensing is performed.","section":"Reference [12]"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is best read as a vision/position paper introducing a new paradigm rather than as a rigorously validated systems paper. The core idea is attractive and likely of interest to the molecular communication community. The main risk is overclaiming experimental validation from uncontrolled re-analysis of communication-testbed data. I would encourage the editor to require the authors to either significantly temper the empirical claims or provide a proper validation with ground-truth airflow and repeated trials. If the journal's scope is strictly archival experimental contributions, the current evidence is insufficient; if it accepts conceptual and exploratory papers, a major revision with corrected framing would be suitable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis paper frames molecular communication (MC) as integrated sensing and communication (ISAC) at macroscale. That framing is the main novelty – using the same received vapor waveform to both decode data and infer airflow, turbulence, or other environmental properties. The authors correctly position it as a complement to EM ISAC in smoke-filled or GNSS-denied environments, and they keep the scope modest (low-rate coordination, ad hoc infrastructure).\n\nWhat’s good: The paper is clearly written and honest. It lays out a system architecture (transmitter, propagation, receiver, joint processing) and lists the obvious open issues – sensor drift, nonlinear response, 3D turbulence, multiple access – in Section VII. The analytical example in Fig. 3 is internally consistent: a fixed detection threshold recovers the OOK bits under two different airflow/attenuation conditions, and the peak shifts would plausibly indicate those conditions. That part works as a simple demonstration of principle.\n\nThe soft spot is the experimental “proof of concept.” Section V reuses data from a prior testbed [14] without running a new controlled experiment. There is no ground-truth airflow measurement for the estimated 0.314 m/s, no control for released ethanol mass or sensor baseline drift, and the MQ-3 sensor’s response time is on the order of seconds – comparable to the observed peak shifts. The 44% earlier / 66% stronger peaks could be explained by a different fan setting, a different spray amount, or humidity. So the statement that environmental parameters are “demonstrated” to be inferred is too strong. The paper’s own Section VII acknowledges exactly these confounds, which makes the overclaim in the abstract and conclusion hard to defend.\n\nThe analytical section is also somewhat self-referential: the same diffusion–advection model generates the waveforms and is used to interpret them. That’s fine for illustration, but it doesn’t show that an independent estimator would work in a real 3D environment.\n\nThe core idea is still sound. Molecular waveforms do contain environmental signatures – that’s not in dispute. The paper’s contribution is framing and a research agenda, not a validated sensing system. Readers interested in MC or ISAC will find the framing useful, but they should read the experimental section skeptically.\n\nI’d send this to peer review, but I’d tell the editors that the experimental claims need substantial revision: either tone them down to “preliminary illustration” or actually run the controlled testbed with ground truth and error bars. The paper is coherent and the authors are not misrepresenting the field; it just overstates its evidence.","headline":"A well-scoped vision paper that overstates the experimental validation for its central sensing claim.","tokens_in":8254,"tokens_out":2685,"would_cite":true,"duration_ms":29251,"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":"VaporISAC claims that chemical vapor pulses can simultaneously transmit information and infer environmental properties such as airflow, turbulence, and smoke, extending ISAC to the molecular domain and demonstrating the concept with analyti","keywords":["molecular communication","integrated sensing and communication","chemical vapor","diffusion-advection channel","airflow sensing","turbulence estimation","on-off keying","MQ-3 sensor"],"falsifier":"In a controlled wind tunnel, release identical OOK vapor pulses at two known airflow velocities while an independent anemometer measures the actual flow; if peak arrival times or amplitudes do not vary in the monotonic way the 1D diffusion-advection model predicts, or if velocities inferred from peak shifts systematically disagree with anemometer readings, the central sensing claim fails.","tokens_in":7418,"feed_emoji":"💨","tokens_out":5137,"duration_ms":50418,"temperature":0.7,"pith_summary":"This paper introduces VaporISAC, a framework that brings integrated sensing and communication (ISAC) into the chemical domain: controlled vapor pulses simultaneously carry data and probe the environment, so that one received waveform yields two outputs—the transmitted message and estimates of airflow, turbulence, smoke, and chemical conditions. The authors argue this matters because EM-based ISAC degrades in smoke-filled, cluttered, enclosed, and infrastructure-denied environments, where molecular propagation through cracks and gaps remains naturally robust. They support the concept with analytical waveforms from a one-dimensional diffusion-advection model and with an experimental ethanol/MQ-3 testbed in which peak shifts and amplitude changes were used to estimate airflow. The paper positions VaporISAC as a complement to, not a replacement for, EM-ISAC, and identifies realistic 3D turbulent channel modeling, chemical multiple access, hardware safety, and validation as open challenges.","feed_headline":"Vapor pulses send data and sense airflow in one waveform","feed_subtitle":"In RF-hostile spaces, one chemical pulse delivers bits and reveals airflow, turbulence, and smoke.","key_machinery":"The central mechanism is the molecular waveform itself, shaped by diffusion and advection: each emitted vapor burst becomes a time-varying concentration profile whose peak arrival time, amplitude, and temporal spreading encode both the transmitted symbol and the environment. The paper calls this a 'one signal, two outputs' paradigm. The analytical demonstrations rely on the one-dimensional diffusion-advection channel model with OOK modulation, a fixed detection threshold for bit recovery, and waveform-feature analysis (peak shift, peak amplitude, spreading/ISI) for sensing. The experimental demonstration uses a spray-based ethanol transmitter and an MQ-3 metal-oxide semiconductor sensor, who","core_discovery":"VaporISAC's central claim is that a chemical vapor pulse can be both a communication symbol and an environmental probe: the same received concentration waveform is jointly processed to recover the transmitted bits and to estimate properties of the propagation environment. The paper argues this extends the ISAC paradigm from electromagnetic signals to molecular signals, where advection and diffusion make the waveform inherently depend on airflow, turbulence, and obstructions. Evidence comes from a one-dimensional diffusion-advection model of an OOK sequence ('101101'), where a fixed detection threshold recovers the bits while peak arrival time indicates airflow velocity and peak amplitude ind","pith_inferences":["If the waveform-to-environment mapping holds beyond the 1D model, the same infrastructure-free vapor channel could double as a distributed airflow and gas sensor network in buildings and ducts; that extension is not in the paper and depends on the 3D turbulence modeling it lists as open.","The paper's reactive-probe idea—vapor molecules that react with ambient chemicals and produce secondary signatures—points toward chemical composition sensing, but it is presented only as a possibility; a concrete test would release a tracer that reacts with a target gas and check whether the secondary signal tracks concentration.","A practical immediate validation would run the same OOK sequence in a wind tunnel with smoke and an independent anemometer: if inferred airflow from peak shifts disagrees with the anemometer under turbulence, the sensing claim needs revision."],"forward_implications":["In environments where RF and LiDAR degrade—smoke, debris, enclosed spaces—molecular waveforms can still carry low-rate messages and infer environmental state from the same received signal.","Sensing requires no dedicated pilot transmissions: every vapor pulse doubles as a probe, and the communication-sensing tradeoff is controlled through symbol rate and observation window.","Peak arrival time can be used to estimate airflow velocity and peak amplitude to estimate propagation losses; the reported experiment yields roughly 0.314 m/s effective drift velocity from a 1 m Tx-Rx separation.","VaporISAC is positioned as a complement to EM-ISAC, motivating hybrid EM-molecular architectures in which each modality handles what the other cannot.","The framework opens applications in GNSS-denied indoor robotics, airflow/duct monitoring, search and rescue in collapsed structures, and coordination of chemical-environment swarms."],"fun_headline_variants":["Chemical pulses: send bits, sense air","One vapor pulse delivers data and airflow","Molecular signals merge comms and sensing","VaporISAC: same wave, two jobs"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is that a one-dimensional diffusion-advection model with constant diffusion and uniform airflow accurately describes real vapor propagation, and that the MQ-3/eNose sensor output tracks normalized concentration; the paper itself lists 3D turbulent propagation, buoyancy, obstacles, thermal gradients, sensor drift, and nonlinearity as open challenges.","fun_headline_variants_meta":{"raw":{"variants":["Chemical pulses: send bits, sense air","One vapor pulse delivers data and airflow","Molecular signals merge comms and sensing","VaporISAC: same wave, two jobs"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000163,"raw_usage":{"total_tokens":1034,"prompt_tokens":656,"completion_tokens":378,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":400,"completion_tokens_details":{"reasoning_tokens":323}},"tokens_in":400,"tokens_out":378,"duration_ms":5170,"temperature":1.0,"reasoning_tokens":323,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T21:52:34.209593+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In a controlled wind tunnel, release identical OOK vapor pulses at two known airflow velocities while an independent anemometer measures the actual flow; if peak arrival times or amplitudes do not vary in the monotonic way the 1D diffusion-advection model predicts, or if velocities inferred from peak shifts systematically disagree with anemometer readings, the central sensing claim fails.","supporting_citations":[],"review_version":1}