{"id":"d59fbd45-5489-4807-a871-d27803645555","arxiv_id":"2411.16501","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A three-antenna 5G testbed estimates Angle of Arrival from Sounding Reference Signals with MUSIC, ESPRIT, and JADE-ESPRIT, reporting under two degrees of error in open-field tests.","lead":"The paper builds a 5G uplink positioning testbed with a three-antenna receiver and tests three Angle of Arrival algorithms on real Sounding Reference Signals. It reports that all three algorithms stay within about two degrees of the true angle in open-field trials, suggesting the testbed works.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The sub-2° accuracy claim depends on an unvalidated phase-calibration chain: no residual phase-error or hold-time test is reported, so an N310 inter-daughterboard phase bias could dominate the reported AoA error.","rationale":"I read the paper as a preliminary testbed paper whose central contribution is the experimental demonstration that the SRS-based pipeline on Ettus N310/E312 can estimate AoA to within 2°. The reader's CONDITIONAL verdict is appropriate. The most load-bearing assumption is the validity of the N310 phase-calibration procedure, because the steering vector in Eq. (3) is the only link between measured phase and angle. The paper describes the calibration in Section 3.1 and Algorithm 1, but does not quantify the residual phase error after real-time compensation. A constant inter-daughterboard phase offset is especially dangerous: with three antennas, MUSIC and ESPRIT have limited degrees of freedom, and even a few degrees of phase error can translate into a bias larger than the claimed accuracy. I agree with the reader's weakest_assumption and propose a concrete calibration-validation experiment to settle it. If the validation passes, the <2° claim is credible; if not, the paper needs to report corrected numbers. No change to the reader's verdict is needed.","tokens_in":8499,"tokens_out":6412,"duration_ms":66004,"concrete_test":"Perform an end-to-end calibration-validation run: with the SRS transmitter off, feed a known CW tone through a matched 1-to-4 splitter into all three (or four) antenna ports of the N310, run Algorithm 1 for 60 s, and record the residual estimated AoA and per-pair phase difference over time. Repeat with the SRS signal on and the transmitter at a mechanically surveyed angle (e.g., using a theodolite or laser-tracking total station). If the residual phase error exceeds about 5° (roughly 0.5–1° in angle for d=λ/2 at 3.5 GHz) or the AoA estimate drifts by more than the claimed 2° across the 60 s window, the current evidence does not support the central claim.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 4.2's claim of 'less than 2 degrees of error' is the load-bearing result. It can only hold if the phase alignment described in Section 3.1 and Algorithm 1 step 7 is correct: the per-pair offsets from the offline 1-to-4 splitter calibration and the real-time differential-LO phase measured on one channel of each daughterboard must both equal the phase offsets on the three antenna channels used for AoA. The paper provides no residual phase-error measurement, no hold-time or thermal-drift test, and no sensitivity analysis. With only M=3 antennas, a constant phase error of, for example, 10° between elements at 3.5 GHz is enough to shift MUSIC/ESPRIT estimates by several degrees, so the reported <2° accuracy is not established. The real-time calibration also shares the calibration tone path with the signal path through a splitter, and any frequency-dependent mismatch between the calibration channel and the antenna channels corrupts the steering-vector model of Eq. (3).","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper describes a software-defined-radio testbed for Angle of Arrival (AoA) estimation using the Sounding Reference Signal (SRS) in 5G NR uplink scenarios. The system uses an Ettus E312 transmitter and an Ettus N310 receiver with a three-element ULA, and the study covers MUSIC, ESPRIT, and JADE-ESPRIT algorithms. The authors present MATLAB ray-tracing simulations for the field environment and report field trials at 2.4 GHz and 3.5 GHz in LOS open-field conditions at 15 m and 50 m. The central claim is that the testbed achieves an accuracy of less than 2 degrees of error, with consistent results over 60-second tests.","tokens_in":8711,"tokens_out":3962,"duration_ms":36779,"significance":"If the performance claim is rigorously established, the paper documents a useful practical platform for single-base-station 5G positioning, including a full SRS-based uplink AoA processing chain and a phase-calibration strategy for a multi-channel USRP. The strengths are the detailed system description, the use of realistic 5G waveforms, and the availability of field measurements rather than only simulations. However, the significance is presently limited by the lack of error bars, statistical aggregation, and calibration validation, which prevents the reader from assessing whether the claimed sub-2-degree accuracy is actually demonstrated.","major_comments":[{"comment":"The central claim of 'less than 2 degrees of error' is not supported by the presented evidence. The paper shows only time-series plots over 60 seconds, without reporting per-test summary statistics such as mean error, standard deviation, RMSE, or bias for each angle, distance, and algorithm. Since each test provides 300 AoA estimates (5 per second over 60 seconds), the authors should provide a statistical table or error bars to substantiate the claimed accuracy and to compare the three algorithms quantitatively.","section":"Section 4.2, Figures 6–8"},{"comment":"The phase calibration procedure is load-bearing for the measured accuracy, but no calibration validation is reported. The offline 1-to-4 splitter calibration and the real-time differential-LO phase correction assume that the phase misalignment remains constant over the 60-second test and that the calibration tone path matches the antenna path. Without a hold-time test, a residual-phase-error measurement, or a sensitivity analysis, a small inter-daughterboard phase bias can directly bias MUSIC/ESPRIT estimates, especially with only M=3 antennas. The authors should quantify the residual phase error after the calibration procedure.","section":"Section 3.1 and Algorithm 1"},{"comment":"The simulation setup is under-specified. The paper does not report the SNR values used in the Monte-Carlo simulations, the number of sources D assumed in MUSIC/ESPRIT/JADE-ESPRIT, or the exact ray-tracing parameters. Without this information, the simulation results in Figures 4 and 5 cannot be reproduced, and the discrepancy with the field results noted in Section 4.2 cannot be evaluated. The authors should state the chosen D and SNR values and provide a direct simulation-measurement comparison under comparable operating conditions.","section":"Section 4.1"},{"comment":"The choice of the number of sources D and the outlier removal rule (3 scaled MAD) are user-defined and not specified for the field tests. The reported accuracy depends on these choices; for instance, an incorrect D in MUSIC can produce false peaks. The authors should state the values used for D across the tests and discuss the sensitivity of the reported accuracy to the outlier threshold.","section":"Section 2.3.2 and Algorithm 1"}],"minor_comments":[{"comment":"The covariance matrix should be E[YY^H] rather than E[YY^T], since the received signal matrix is complex and the Hermitian transpose is required for a valid covariance matrix.","section":"Eq. (6)"},{"comment":"The summation index in the cross-correlation runs from i=1 to Lseq-1; please verify whether the index should start at i=0 or whether the upper limit is intended to be Lseq.","section":"Eq. (5)"},{"comment":"The legend text '2D ESPRITMUSICESPRIT' lacks spacing between the algorithm names; it should read '2D ESPRIT, MUSIC, ESPRIT'.","section":"Figure 6 caption"},{"comment":"The text states that the angle of arrival for the 2.4 GHz tests was fixed at 0 degrees, but Figures 7 and 8 show estimates around 1.5 to 2 degrees and -1 to -2 degrees, respectively; the authors should quantify the precision of the ground-truth angle setup and the positional alignment uncertainty.","section":"Section 4.2"},{"comment":"The statement that 'the discrepancy between the simulation results and actual measurements may seem significant' is vague; please provide a quantitative comparison of simulated versus measured error statistics to make the discussion concrete.","section":"Section 4.2"}],"recommendation":"major_revision","confidential_remarks":"This is a work-in-progress workshop paper. The main load-bearing issue is the unvalidated calibration chain and the absence of statistical error characterization. Both are fixable within the manuscript's scope by adding a hold-time test, residual phase error analysis, and per-test error tables. The paper is within the venue's scope, and the testbed description is valuable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know this is a work-in-progress testbed paper, not a new algorithm paper. The authors built a 5G uplink AoA testbed on an Ettus N310/E312 pair, with a 3-element ULA, and ran MUSIC, ESPRIT, and JADE-ESPRIT on real SRS signals in an open field. That is the actual contribution: the specific hardware chain, the calibration procedure, and the field measurements. The algorithms themselves are textbook, and prior work has already done AoA with USRPs, so the novelty is the 5G SRS implementation and the calibration pipeline. That is a legitimate extension, and the paper is honest about being preliminary.\n\nWhat the paper does well: the testbed description is detailed, the calibration problem (two daughterboards with independent LOs, phase misalignment) is addressed head-on, and the field trials cover multiple angles and distances. The ray-tracing simulation model is a sensible way to support the design. The time-series plots show the three algorithms tracking the commanded angles over 60 seconds, which is decent evidence that the pipeline works.\n\nNow the soft spots. The load-bearing claim of \"less than 2 degrees of error\" is not actually backed by the plots—Figure 7 shows deviations up to about 3 degrees for a 0° target, and Figure 6 shows noticeable spikes at 0°, 10°, and 15°. There are no error bars, no ground-truth angle uncertainty, no aggregate statistics beyond a single-sentence claim. The number of sources D used in the algorithms is never stated, nor is the simulation SNR. That matters because with only 3 antennas, D is critical and MUSIC/ESPRIT are sensitive to phase errors at 3.5 GHz.\n\nThe more serious concern, which the stress-test note raises, is the calibration validity. The paper describes a two-step phase alignment: an offline splitter calibration for channel pairs and a real-time differential-LO measurement on one channel of each pair. But there is no residual phase-error measurement, no hold-time or drift test, and no sensitivity analysis. If the calibration tone path is frequency-dependent or the LO phase drifts, the steering-vector model breaks and the AoA estimates become biased. The paper needs to show that the calibration actually holds during the measurement window.\n\nAll that said, the central argument—that the testbed can estimate AoA with SRS in LOS conditions—is plausible and supported by the data the authors did show. It is not a load-bearing flaw, but it is a missing-validation problem. The paper deserves a serious referee, but the referee should require the missing parameters, error statistics, and a calibration-validity test before publication.\n\nWho is this for? Researchers working on practical SDR-based positioning testbeds, especially for 5G NR, will find value here. It is a niche result, but a useful data point. I would encourage you to engage with it if you are in that space; otherwise, it is a pass.","headline":"A useful testbed paper with honest limitations; the <2° accuracy claim needs phase-calibration validation and proper error statistics.","tokens_in":9263,"tokens_out":1496,"would_cite":false,"duration_ms":15486,"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 claims that a single 5G base station with a three-element antenna array can estimate the angle of arrival of an uplink user's Sounding Reference Signal with less than 2 degrees of error in open-field line-of-sight conditions…","keywords":["Angle of Arrival","5G NR","Sounding Reference Signal","MUSIC","ESPRIT","JADE-ESPRIT","Uniform Linear Array","Software Defined Radio"],"falsifier":"Run a static 0-degree test at 15 meters while time-stamping the calibration tone's differential phase every second for 60 seconds; if the angle estimate moves by more than 2 degrees when the correction is held fixed, or if the residual phase changes by more than the equivalent of 2 degrees of angle of arrival, the accuracy claim collapses. Alternatively, compare field estimates against a surveyed ground-truth angle from a theodolite or RTK position.","tokens_in":8334,"feed_emoji":"📡","tokens_out":5357,"duration_ms":44873,"temperature":0.7,"pith_summary":"This paper claims that a single 5G base station, equipped with a three-element uniform linear array, can estimate the angle of arrival of a user's uplink Sounding Reference Signal with an error below 2 degrees in open-field line-of-sight conditions. The claim is supported by ray-tracing simulations of the test environment and by field trials at 15 and 50 meters in the 2.4 GHz and 3.5 GHz bands. The authors compare three super-resolution algorithms, MUSIC, ESPRIT, and JADE-ESPRIT, and find comparable performance across tested angles from 0 to 45 degrees. If accurate, this makes single-node positioning in GNSS-denied areas a practical option with commercial software-defined radios rather than specialized hardware.","feed_headline":"SRS-based 5G testbed estimates angle of arrival within 2 degrees","feed_subtitle":"Open-field MUSIC, ESPRIT, and JADE-ESPRIT trials track a transmitter to sub-2° accuracy at 15 and 50 m.","key_machinery":"The load-bearing machinery is the three-element uniform linear array phase-aligned through a two-stage calibration: offline per-pair phase offsets from a splitter-injected tone, and real-time inter-pair alignment using a common signal, as described in Algorithm 1. On this aligned array, the SRS pilots are extracted after cross-correlation timing synchronization, and each of MUSIC, ESPRIT, and JADE-ESPRIT is applied to the frequency-domain snapshot to produce one angle estimate per second. The steering-vector relation in Eq. (3) is what connects the measured inter-antenna phase differences to the azimuth angle.","core_discovery":"The paper's central discovery is that a phase-calibrated software-defined radio receiver, using only three of its four channels, can estimate the angle of arrival from 5G SRS pilots with consistent sub-2-degree accuracy over a 60-second static test. The authors attribute this to compensating for the receiver's two daughterboard local oscillators by injecting a common signal and measuring the differential phase in real time, combined with the known steering-vector model in Eq. (3). Simulations with a ray-tracing model of the same field, including a single ground reflection and a nearby building, show all three algorithms behaving similarly, and the field tests confirm this pattern.","pith_inferences":["The 2-degree figure is reported for static, open-field line-of-sight with a single dominant path; in dense multipath or with moving users the accuracy would likely degrade, since three antennas can resolve at most two sources.","A fair test of the calibration assumption would be to repeat a 0-degree run while logging the estimated phase offset every second; if the offset drifts, the central accuracy claim would need to be re-baselined.","The paper's alignment uncertainty from manual transmitter placement could be separated from estimator error by using a surveyed ground-truth position, such as from RTK GPS, as reference instead of the nominal angle.","The ray-tracing model with one-bounce reflections could be extended to include diffuse scattering or vegetation to see how robust the sub-2-degree performance is in less ideal open fields."],"forward_implications":["A 5G base station with just three antenna elements can localize an uplink user in angle well enough for single-node positioning in open environments.","The existing 5G Sounding Reference Signal can serve as the positioning signal, so operators do not need a separate reference signal or dedicated hardware.","The comparable performance of MUSIC, ESPRIT, and JADE-ESPRIT means a simpler closed-form estimator such as ESPRIT may be sufficient in line-of-sight conditions.","The same testbed can be extended to higher bands such as 5.8 GHz, as the authors state is their ongoing work.","Real-time phase calibration between daughterboard channel pairs is a practical solution for coherent processing on multi-channel software-defined radios."],"supporting_citations":[{"why":"Provides the single-base-station ToA-AoA localization testbed that this work builds upon.","marker":"[1]"},{"why":"Compares 3D MUSIC with expectation-maximization for single-station positioning, setting the context for the algorithms used here.","marker":"[2]"},{"why":"Defines the 5G NR physical layer, including SRS generation and subcarrier mapping, which the transmitted waveforms follow.","marker":"[6]"},{"why":"Supplies the ray-tracing propagation modeling method used for the simulation environment.","marker":"[8]"},{"why":"Original MUSIC algorithm, one of the three estimators whose performance is evaluated.","marker":"[9]"},{"why":"Original ESPRIT algorithm with rotational invariance, the second estimator evaluated.","marker":"[10]"},{"why":"Introduces joint angle and delay estimation via shift-invariance, the basis for JADE-ESPRIT.","marker":"[11]"},{"why":"Documents the phase misalignment between daughterboard channel pairs in the N310 receiver, motivating the real-time calibration step.","marker":"[13]"}],"fun_headline_variants":["Sub-2° angle of arrival from 5G SRS on a 3-element array","5G SRS positioning: three channels, sub-2° accuracy","Practical 5G AoA: sub-2° from SRS on a ULA","Three-antenna SRS receiver hits sub-2° AoA in field tests"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole measurement chain assumes the phase offsets measured during calibration stay fixed for the 60 seconds of each test; if either local oscillator drifts, the steering-vector model is wrong and every angle estimate becomes biased.","fun_headline_variants_meta":{"raw":{"variants":["Sub-2° angle of arrival from 5G SRS on a 3-element array","5G SRS positioning: three channels, sub-2° accuracy","Practical 5G AoA: sub-2° from SRS on a ULA","Three-antenna SRS receiver hits sub-2° AoA in field tests"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000516,"raw_usage":{"total_tokens":2482,"prompt_tokens":899,"completion_tokens":1583,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":515,"completion_tokens_details":{"reasoning_tokens":1501}},"tokens_in":515,"tokens_out":1583,"duration_ms":11798,"temperature":1.0,"reasoning_tokens":1501,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T13:02:10.946068+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a static 0-degree test at 15 meters while time-stamping the calibration tone's differential phase every second for 60 seconds; if the angle estimate moves by more than 2 degrees when the correction is held fixed, or if the residual phase changes by more than the equivalent of 2 degrees of angle of arrival, the accuracy claim collapses. Alternatively, compare field estimates against a surveyed ground-truth angle from a theodolite or RTK position.","supporting_citations":[{"cited_title":"Blanco, N","cited_arxiv_id":null,"evidence_quote":"Provides the single-base-station ToA-AoA localization testbed that this work builds upon."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Compares 3D MUSIC with expectation-maximization for single-station positioning, setting the context for the algorithms used here."},{"cited_title":"Physical Channels and Modulation","cited_arxiv_id":null,"evidence_quote":"Defines the 5G NR physical layer, including SRS generation and subcarrier mapping, which the transmitted waveforms follow."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Original ESPRIT algorithm with rotational invariance, the second estimator evaluated."},{"cited_title":"van der Veen, M","cited_arxiv_id":null,"evidence_quote":"Introduces joint angle and delay estimation via shift-invariance, the basis for JADE-ESPRIT."}],"review_version":1}