{"id":"199798ff-7362-4fdd-9d2a-27456a2ac895","arxiv_id":"2501.13144","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"GWEn is an open-source, ~300 euro, two-axis rotating testbed that kept UWB mean-distance measurements reproducible within 2.3 cm and added less than 1.7 cm of measured interference, below the device's own +/-10 cm accuracy.","lead":"The paper introduces GWEn, a roughly 300 euro, 3D-printed and open-source two-axis rotating stand that moves wireless devices like smartphones to measure how their antennas behave in different directions. A smart generalist might care because cheap, portable, reproducible wireless test gear could make antenna and signal measurements much more accessible outside expensive anechoic chambers.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The most load-bearing gap is that minimal RF interference is demonstrated only as UWB time-of-flight through the tower, leaving the arm, holding plate, and USB cable untested and excluding amplitude and phase; the general proven-testbed claim overreaches.","rationale":"The reader's weakest assumption matches my reading almost exactly, so I agree rather than manufacture a new objection. What I add is quantitative sharpening: the paper's own measured UWB error is small in distance but large in phase. That makes the missing amplitude and phase test decisive for the paper's stated use case of radiation-pattern evaluation. I do not see grounds for rejection: the overgeneralization is a scope problem, not an internal inconsistency, and the open-source release is a real artifact that can be checked. Keeping the CONDITIONAL verdict is correct; the fix is to require either a VNA-based S21 evaluation or a scoping of the interference claim to UWB time-of-flight applications.","tokens_in":892,"tokens_out":1505,"duration_ms":78621,"concrete_test":"Run a VNA-based S21 measurement: mount a known reference antenna on GWEn exactly as in the workflow, with the USB cable connected, and compare magnitude and phase against the same antenna on an RF-transparent foam stand at the same location. Sweep 2-8 GHz and, if mmWave support is claimed, at 24 and 60 GHz; repeat at representative theta and phi positions such as 0, 90, 180, and 270 degrees on both axes. If the GWEn-mounted pattern differs from the control by more than about 0.5 dB in gain or 10 degrees in phase at the frequencies of interest, the minimal RF interference claim must be restricted to UWB time-of-flight.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that GWEn is a reproducible and proven testbed with minimal influence on wireless signals is supported by exactly one quantitative interference experiment (Section 4.1). That experiment compares UWB time-of-flight distance with and without the tower between two DWM3000EVB boards. It therefore bounds only the group-delay effect of the tower on one UWB link. It does not cover the rotation arm and holding plate, which are the structures nearest the DUT antenna and which move with it; the USB cable required by the workflow (Section 3, step 2); or the metallic motor and ball bearings inside the tower. More fundamentally, the selected metric is distance, not the amplitude or phase that define antenna radiation patterns. Even the reported 1.684 cm mean distance increase corresponds to roughly 160 degrees of phase at 8 GHz and several full wavelengths at the mmWave frequencies claimed in Section 5.2. Hence the Section 2.1 requirement of little interference to electromagnetic signals is not demonstrated for the intended radiation-pattern use case, and the abstract promise of mmWave and acoustic support is not backed by RF-transparency data in those bands. This is a scope gap, not an internal inconsistency: the paper carefully calls the UWB measurement exemplary, but the general conclusion in Section 7 goes beyond what the data can prove.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents GWEn, an open-source, 3D-printed, two-axis gimbal platform for wireless physical-layer evaluation. It describes the mechanical design, the software/web workflow, a seven-step measurement procedure, and an evaluation consisting of two experiments: (i) an interference test that compares UWB time-of-flight distance measurements with and without the tower placed between two DWM3000EVB boards, and (ii) a reproducibility test that repeats a 360-degree base rotation three times in succession and once after one week. The paper reports a mean distance increase of 1.684 cm when the tower blocks the line of sight, below the manufacturer's +/-10 cm UWB accuracy, and mean-distance differences of at most 0.632 cm within a day and 2.264 cm after one week. It also sketches UWB, mmWave, and acoustic use cases and concludes that GWEn is a reproducible and proven testbed with minimal influence on wireless signals.","tokens_in":9892,"tokens_out":10799,"duration_ms":112300,"significance":"If the claims hold, GWEn is a valuable low-cost (~300 EUR), portable (<5 kg) open-source positioner that could lower the barrier to physical-layer measurements. The paper's strengths are its concrete open-source release, the clear measurement workflow, the large sample sizes (10,000 UWB packets per interference condition), and the honest use of external datasheet and permittivity references rather than fitted parameters. The main limitation is that the 'minimal RF interference' conclusion rests on a single UWB time-of-flight metric through the tower only; the amplitude and phase quantities that matter for radiation-pattern measurements are not evaluated, and components near the DUT (rotation arm, holding plate, USB cable) are not tested. The reproducibility evidence is based on orientation-averaged means and would benefit from per-angle agreement metrics.","major_comments":[{"comment":"The interference evaluation measures only UWB time-of-flight distance with the tower placed between two DWM3000EVB boards. This does not substantiate the Section 2.1 design requirement of 'little interference to electromagnetic signals' for the intended radiation-pattern use case, which depends on amplitude and phase, not only group delay. The reported 1.684 cm mean distance shift is about 0.45 wavelengths at 8 GHz (roughly 160 degrees of phase) and several wavelengths at the mmWave frequencies claimed in Section 5.2, so being below the +/-10 cm UWB ranging accuracy is not evidence of small phase or amplitude perturbation. Please add an S-parameter or amplitude/phase pattern measurement covering the rotation arm, holding plate, and USB cable used in the Section 3 workflow, or explicitly limit the interference claim to UWB ranging.","section":"Section 4.1"},{"comment":"The tested configuration does not cover the components closest to the DUT antenna. In the LoS/NLoS comparison, the tower is between the two devices, while the rotation arm and holding plate sit immediately next to the DUT and rotate with it, and the standard workflow (Section 3, step 2) requires a USB cable attached to the DUT. These elements are absent from the described test. A nearby cable or structure can alter the antenna's effective radiation pattern and impedance, so the measured 1.684 cm bound cannot be generalized to the full GWEn configuration. In addition, the description does not make clear whether the metal motor and ball bearings were in the direct path, so the phrase 'maximum introduced error' is not justified.","section":"Section 4.1"},{"comment":"The conclusion that 'GWEn represents a reproducible and proven testbed' goes beyond the evidence in Section 4. Only one device pair, one geometry, and one metric were evaluated. If additional experiments are not possible, the conclusion should be qualified to 'initial evidence' or 'for UWB ranging', and the Section 4.1 claim of identifying the maximum introduced error should be softened.","section":"Section 7"},{"comment":"The reproducibility conclusion is based on the mean distance aggregated over all orientations, while the reported standard deviations are about 11 cm, roughly 22% of the 50 cm nominal distance. The per-angle plots in Figure 5 are qualitative; no per-orientation agreement metric such as mean absolute angular error or RMS pattern difference is reported. To support the 'reproducible measurements' claim for radiation-pattern use, please provide a quantitative per-angle comparison and state whether the standard deviation in Table 1 is per-sample or per-position.","section":"Section 4.2 / Table 1"}],"minor_comments":[{"comment":"The sentence 'the only part which could interfere with signals is the hollow tower' is contradicted by the following sentence acknowledging that metal-based motors and ball bearings can influence or block the entire signal; please reconcile these statements.","section":"Section 4.1"},{"comment":"Sections 5.2 and 5.3 describe potential use cases rather than measurements, but Section 7 says three distinct use cases were 'showcased' and 'applied successfully'; only UWB has experimental data (Figure 6). Please revise the wording or add pilot results.","section":"Section 5.2 / 5.3"},{"comment":"The manual adjustment of the tower and holding plate to place the DUT antenna exactly at the intersection of the two rotation axes is described, but no verification procedure or tolerance analysis is given; a short description of how to check alignment would improve reproducibility.","section":"Section 2.1 / Section 3"},{"comment":"The measured mean distances in Table 1 are consistently below the nominal 50 cm shown as the dotted line in Figure 5, but the paper does not discuss this offset; please clarify whether it is an expected UWB bias or a setup artifact.","section":"Section 4.2 / Figure 5"},{"comment":"There are minor typographical inconsistencies, including 'GWEN' in Section 5.1 and the abstract's line break 'measuringan-tenna'; these should be corrected.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"This is a systems/tool paper, and the open-source release, workflow, and reproducibility experiment are genuine contributions. The main risk is that the 'minimal RF interference' and 'proven testbed' claims are broader than the single UWB time-of-flight experiment in Section 4.1. I would be comfortable with a major revision if the authors either add a small amplitude/phase experiment covering the arm, holding plate, and cable, or carefully restrict all interference and generality claims to UWB ranging. I did not find evidence of circularity or fabricated parameters; the benchmark against the datasheet and the cited permittivity value are appropriate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"GWEn is a real, usable piece of hardware engineering: a ~300 EUR, 3D-printed, two-axis gimbal with an open-source control stack and a sensible workflow for rotating a smartphone or other DUT around a fixed center point. The quantitative evaluation is honest as far as it goes: a UWB time-of-flight test shows the tower induces a 1.684 cm mean distance shift, below the DW3000's ±10 cm rated accuracy, and three consecutive runs plus a one-week-later repeat stay within ~2.3 cm of each other. That is credible evidence that the platform is reproducible for UWB ranging.\n\nThe soft spots are real but not fatal. The 'minimal RF interference' claim is established only for one geometry: the hollow tower directly between two UWB transceivers, measuring distance. It does not cover the rotation arm and holding plate that sit closest to the DUT antenna, the USB cable the workflow requires, or the metallic motor and bearings. More importantly, distance is not the metric that defines an antenna radiation pattern; amplitude and phase are. The paper acknowledges the UWB test is exemplary, but then the conclusion calls GWEn a 'proven testbed' with 'minimal influence on wireless signals'—that is an overreach. The 0.01° positioning resolution is theoretical and unverified, and the mmWave and acoustic sections describe future use cases, not measurements. These are scope gaps, not internal contradictions. The self-citation to the authors' UWB accuracy preprint is benign. The related-work discussion is fair and explicitly positions GWEn against prior open-source rigs, including the similar Raspberry-Pi-based one.\n\nWho should read it: anyone building low-cost wireless measurement setups, especially for UWB or directionality studies, will get value from the design, the part list, and the workflow. It is not a contribution to RF metrology or antenna theory. I would send it to peer review with a request that the authors either add RF-transparency measurements for the near-field components (arm, holding plate, cable) and at least one amplitude/phase sweep, or soften the general claims to 'minimal influence on UWB ranging.' The paper deserves a serious referee, but the referee should hold the authors to that.","headline":"A genuinely useful open-source positioner for UWB experiments, but its general 'minimal RF interference' claim is under-supported; referee it with a request for broader RF transparency data.","tokens_in":10521,"tokens_out":2254,"would_cite":true,"duration_ms":23059,"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 presents GWEn, an open-source 3D-printed two-axis positioner whose measured UWB distance error is 1.684 cm, well below the device's advertised ±10 cm accuracy, and whose repeated 360° runs reproduce mean distances within 0.632…","keywords":["wireless physical layer","antenna radiation patterns","open-source testbed","ultra-wideband","UWB ranging","3D-printed positioner","reproducible measurements","acoustic communication"],"falsifier":"Measure the same antenna's radiation pattern in an anechoic chamber once with the full GWEn assembly (including the rotation arm and a USB cable attached) and once with the antenna on a non-conductive foam stand; if the arm or cable produces pattern deviations beyond the tolerance implied by the tower-only UWB test, the minimal-interference claim fails.","tokens_in":9403,"feed_emoji":"📡","tokens_out":14793,"duration_ms":126872,"temperature":0.7,"pith_summary":"GWEn is a lightweight (under 5 kg), compact (50 cm × 50 cm), 3D-printed two-axis positioner that rotates a wireless device around a fixed center point to evaluate antenna radiation patterns and physical-layer behavior. The paper's central claim is that GWEn causes minimal radio-frequency interference: in a UWB ranging test, placing the tower between two devices increased the measured distance by 1.684 cm, which is well below the ±10 cm accuracy the UWB transceiver advertises. It also claims reproducible measurements: three consecutive 360° runs differed by at most 0.632 cm, and a rerun one week later differed by 2.264 cm. If true, GWEn offers a ~€300 open-source alternative to commercial multi-axis positioners, bringing portable, automated physical-layer evaluation to researchers. The paper additionally documents a measurement workflow and use cases for UWB, mmWave, and acoustic communication.","feed_headline":"Open-source 3D-printed rig adds 1.7 cm of error to UWB ranging tests","feed_subtitle":"An open-source €300 positioner brings reproducible wireless testing out of the anechoic chamber.","key_machinery":"The central object is GWEn itself: a two-axis gimbal positioner built from 3D-printed PLA parts and off-the-shelf components, with the device under test mounted on a rotation arm whose holding plate can be adjusted so the DUT's antenna sits exactly at the center point where the two rotation axes meet — the arm's x-axis, angle $\\phi$, and the base's y-axis, angle $\\theta$. Each axis rotates with a theoretical resolution of 0.01°, and the entire assembly is controlled by a Raspberry Pi with a web interface; a USB cable carries measurement data from the DUT to the controller. The argument for minimal interference rests on the low relative permittivity of PLA (2.52, from a cited characterization) and on UWB time-of-flight ranging, which detects added propagation delay as an increased measured distance. The reproducibility claim is carried by the automated rotation routine, which steps the DUT through a full 360° of $\\theta$ at fixed $\\phi$ and records measurements at every position.","core_discovery":"On its own terms, GWEn demonstrates that a low-cost, 3D-printed positioner can serve as a reproducible testbed for wireless physical-layer evaluation. The load-bearing quantitative evidence comes from UWB ranging: with 10,000 measurements per configuration, the mean distance in clear line of sight was 34.660 cm (standard deviation 1.345 cm), and with the hollow PLA tower interposed between the two UWB transceivers it rose to 36.344 cm (standard deviation 1.735 cm), an added error of 1.684 cm, which the authors describe as well below the ±10 cm accuracy margin of the Qorvo DW3000 device. Reproducibility is supported by three consecutive automated 360° runs whose mean distances stayed within 0.632 cm, and a fourth run one week later within 2.264 cm. From this the authors conclude that GWEn is capable of conducting reproducible measurements with minimal influence on wireless signals, and they position it as an extensible platform for UWB, mmWave, and acoustic physical layers.","pith_inferences":["Beyond the paper: because the interference test only places the hollow tower between two UWB devices, the rotation arm, holding plate, and required USB cable are never measured; if they detune the antenna, radiation-pattern measurements could carry more error than the UWB distance figure suggests.","Beyond the paper: a network-analyzer or channel-sounder measurement of the full assembly would convert the 1.684 cm distance error into the amplitude and phase distortions that actually matter for antenna pattern work.","Beyond the paper: extending the same reproducibility protocol to mmWave and acoustic links would test whether the platform's low-interference claim holds outside UWB, where the paper's evidence currently concentrates."],"forward_implications":["Researchers can build a ~€300 open-source positioner for antenna radiation-pattern measurements without an anechoic chamber, following the released 3D files and assembly instructions.","Automated UWB evaluations of real-world devices such as smartphones become repeatable across orientations, environments, and locations, supporting work on digital car key security.","GWEn's extensible architecture — plugin sources, a Python web interface, and an optional linear rail — allows new DUTs and movement patterns to be added with modest effort, covering mmWave beam-pattern characterization and acoustic data-over-sound measurements.","The stored per-angle measurement data and initial configuration ZIP file give researchers a reproducibility trail for physical-layer experiments, including recovery after power interruption.","Because the tower-induced error (1.684 cm) is under the UWB device's ±10 cm accuracy, GWEn can be used for ranging experiments where the platform itself does not dominate the error budget."],"supporting_citations":[{"why":"Supplies the measured relative permittivity of PLA (2.52) used to argue that GWEn's 3D-printed structure causes only minor signal interference.","marker":"[4]"},{"why":"Sets the DW3000 device's advertised ±10 cm ranging accuracy that the 1.684 cm tower-induced error is compared against.","marker":"[24]"},{"why":"Provides the ~USD 3,000 academic low-cost positioner that GWEn's ~€300 material cost is compared with to establish affordability.","marker":"[11]"},{"why":"Describes an existing 3D-printed Raspberry Pi-based 2-axis rotation system, the prior state of the art that GWEn extends with a full PLA enclosure.","marker":"[9]"},{"why":"Supplies the open-source C-Beam linear rail that GWEn mounts on, supporting the claimed extensibility for automated distance-varied measurements.","marker":"[20]"},{"why":"Presents the authors' earlier UWB smartphone-ranging results that GWEn's current UWB measurement workflow builds on.","marker":"[14]"},{"why":"Documents a practical UWB distance-reduction attack, motivating automated evaluation of UWB physical layers with a portable positioner.","marker":"[18]"}],"fun_headline_variants":["Low-cost open-source rig evaluates wireless signals on the move","GWEn makes wireless testing accessible with 3D-printed parts","Open-source GWEn rig tests UWB, mmWave, and acoustic in field","Portable 3D-printed positioner for reproducible wireless tests","GWEn: €300 open-source platform for wireless physical-layer tests"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim that GWEn minimizes wireless interference rests on a single UWB experiment in Section 4.1 that places only the hollow tower between the two devices, while the rotation arm, holding plate, and USB cable that sit closest to the antenna are never measured for their effect.","fun_headline_variants_meta":{"raw":{"variants":["Low-cost open-source rig evaluates wireless signals on the move","GWEn makes wireless testing accessible with 3D-printed parts","Open-source GWEn rig tests UWB, mmWave, and acoustic in field","Portable 3D-printed positioner for reproducible wireless tests","GWEn: €300 open-source platform for wireless physical-layer tests"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000508,"raw_usage":{"total_tokens":2457,"prompt_tokens":907,"completion_tokens":1550,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":523,"completion_tokens_details":{"reasoning_tokens":1458}},"tokens_in":523,"tokens_out":1550,"duration_ms":14535,"temperature":1.0,"reasoning_tokens":1458,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T16:33:45.760012+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the same antenna's radiation pattern in an anechoic chamber once with the full GWEn assembly (including the rotation arm and a USB cable attached) and once with the antenna on a non-conductive foam stand; if the arm or cable produces pattern deviations beyond the tolerance implied by the tower-only UWB test, the minimal-interference claim fails.","supporting_citations":[{"cited_title":"Boussatour, P.-Y","cited_arxiv_id":null,"evidence_quote":"Supplies the measured relative permittivity of PLA (2.52) used to argue that GWEn's 3D-printed structure causes only minor signal interference."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Sets the DW3000 device's advertised ±10 cm ranging accuracy that the 1.684 cm tower-induced error is compared against."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the ~USD 3,000 academic low-cost positioner that GWEn's ~€300 material cost is compared with to establish affordability."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Describes an existing 3D-printed Raspberry Pi-based 2-axis rotation system, the prior state of the art that GWEn extends with a full PLA enclosure."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the open-source C-Beam linear rail that GWEn mounts on, supporting the claimed extensibility for automated distance-varied measurements."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Documents a practical UWB distance-reduction attack, motivating automated evaluation of UWB physical layers with a portable positioner."}],"review_version":1}