Pith. sign in

REVIEW 3 major objections 6 minor 1 cited by

An Experimental Multi-Band Channel Characterization in the Upper Mid-Band

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that in the low FR3 band, adding a reflective target increases the number of resolvable multipath components at 6.5 GHz and decreases it at 8.75 GHz.

desk verdict Useful first FR3 multi-band ISAC measurements, but the central mode-shift claim depends on an uncalibrated estimator and needs synthetic validation. read the letter →

arxiv 2411.12888 v1 pith:GYX4E7JE submitted 2024-11-19 cs.IT eess.SPmath.IT

Roberto Bomfin , Ahmad Bazzi , Hao Guo , Hyeongtaek Lee , Marco Mezzavilla , Sundeep Rangan , Junil Choi , Marwa Chafii This is my paper · ORCID
classification cs.ITeess.SPmath.IT
keywords FR3uppermid-bandchannelmodelingmulti-bandmeasurementsintegratedsensingandcommunicationMUSICdelayestimationmultipathcomponentsindoorpropagation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper reports a two-frequency indoor channel measurement campaign in the low FR3 band, at 6.5 GHz and 8.75 GHz, designed to see how the presence of a reflective target changes the multipath channel. The central claim is that the target affects the two frequencies in opposite ways: at 6.5 GHz new distinguishable reflections appear, while at 8.75 GHz existing paths are blocked. This matters for integrated sensing and communication because it suggests that the best sensing strategy depends on which FR3 frequency is used: lower frequencies can exploit newly created late-arriving paths, while higher frequencies should detect targets by looking for path suppression. The paper also offers a processing methodology, based on MUSIC delay refinement with frequency smoothing and clustering, as a benchmark for future FR3 multi-band studies.

What carries the argument

The central machinery is a frequency-smoothed MUSIC power-delay-profile estimator. Because multipath components are coherent, the channel covariance is rank-one; the paper applies frequency-domain smoothing to restore rank, uses a second-order elbow method on the smoothed eigenvalues to select the number of paths, and then evaluates a MUSIC pseudo-spectrum to refine delays. K-means clustering with a silhouette criterion groups the per-frame delay estimates into stable clusters. The path-count histograms and the positive/negative region analysis are all computed from these estimates, so the entire frequency-dependent conclusion rests on this pipeline.

What would settle it

Run the same measurement and processing pipeline on synthetic channels with a known number of multipath components, matched to the SNR and bandwidth of both frequency bands, with and without a simulated target. If the estimated path-count modes shift in the same direction as the measurements despite no physical target-induced path creation or blockage, the reported frequency dependence is an artifact of the estimator.

Watch

Extended reading notes

Core claim

On its own terms, the paper claims that the target's effect on the background channel is frequency-dependent. At 6.5 GHz, histograms of the estimated number of multipath components shift to higher counts when the target is present, such as a mode increasing from 3 to 4 across all orientations. At 8.75 GHz, the mode shifts to lower counts, from 4 to 3 across all orientations, and from 4 to 2 for one orientation. The paper interprets this as lower frequencies diffracting around the target to create new multi-bounce paths, while higher frequencies are more susceptible to blockage and reflection losses. It further identifies positive regions, where new target-related reflections appear, and negative regions, where the target suppresses existing clutter paths, and argues that each supports a distinct sensing modality.

Load-bearing premise

The whole frequency-dependent pattern depends on the second-order elbow method correctly counting the number of multipath components from the smoothed covariance matrix at both frequencies, with and without the target; if that estimator shifts with SNR or eigenvalue structure, the reported mode changes could be numerical artifacts rather than propagation effects.

Editorial extensions

If this is right

  • At 6.5 GHz, a sensing receiver should expect additional late-arriving multipath components created by the target; these positive-region paths carry target information without overlapping main clutter.
  • At 8.75 GHz, target presence is better detected as a reduction in the number of paths; negative-region blockage analysis is the more promising sensing mode.
  • The two frequencies cannot share a single multipath-count model, so FR3 channel models and ISAC algorithms should treat sub-bands separately.
  • The methodology, combining smoothed MUSIC with elbow-based model order selection and K-means clustering, gives a repeatable benchmark for comparing future FR3 multi-band measurements.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A testable extension would be to sweep intermediate frequencies between 6.5 and 8.75 GHz to locate where the target's effect switches from path creation to path suppression.
  • Because the result is derived from an uncalibrated model-order estimator, an SNR-matched synthetic-channel experiment is needed to separate physical propagation effects from estimator artifacts.
  • If the pattern holds, a dual-band ISAC system could fuse the two modalities: use the lower band's new reflections for detection and the higher band's suppression for localization or shadow estimation.
  • The observed stability of the 8.75 GHz delay clusters hints that higher FR3 frequencies may provide more precise delay estimation once blockage is modeled, a hypothesis worth testing with a larger antenna array.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. This paper reports on an indoor multi-band channel measurement campaign at 6.5 GHz and 8.75 GHz, conducted with and without a metallic target in the environment, using the Pi-Radio SDR platform. The processing chain consists of frequency-domain smoothing, a second-order elbow method for model-order selection, and MUSIC-based delay estimation, followed by K-means clustering of delay estimates and an illustrative positive/negative region analysis. The central empirical claim is that the presence of the target increases the number of distinguishable multipath components at 6.5 GHz, while at 8.75 GHz it reduces that number, which is interpreted as frequency-dependent blockage.

Significance. If the frequency-dependent effect is real, the finding would be a useful input to ISAC design in FR3, suggesting that lower FR3 frequencies can exploit newly created late reflections while higher FR3 frequencies are better served by detecting clutter suppression. The paper provides a valuable measurement setup description and a novel dataset; the authors are also transparent about the antenna being used outside its rated band (Section IV-A). However, the strength of the evidence is currently insufficient: the central claim is read from histograms of a model-order estimate whose bias is not calibrated, and there is no statistical uncertainty quantification or validation on synthetic channels. These issues are addressable and do not invalidate the potential value of the measurements, but they must be resolved before the claim is accepted as established.

major comments (3)
  1. [Section III-A and Section V-B] The second-order elbow method used to estimate the number of multipath components, bLa, is not validated or calibrated. The central result—the histogram mode shifts in Figs. 4 and 5—depends entirely on this estimator being a condition-invariant proxy for the number of distinguishable paths. The measurements include a frequency-dependent SNR difference: the antenna is used at 8.75 GHz outside its specified 6.0–8.5 GHz band (Section IV-A), and target presence changes path amplitudes. An uncalibrated elbow method can systematically undercount at lower SNR or when late paths are weak, which would produce the observed reduction at 8.75 GHz and the apparent increase at 6.5 GHz as a numerical artifact. Please add validation on synthetic channels with known numbers of MPCs, including the SNR and eigenvalue regimes of the measurements, or otherwise demonstrate that bLa is not biased differentially across frequency and target conditions.
  2. [Section V-B] The reported mode shifts are not accompanied by any uncertainty quantification or statistical test. The histograms in Figs. 4 and 5 have broad, overlapping distributions; for example, in Fig. 5 (Beta), the 'no target' distribution has many counts at 4 and 5, while the 'with target' distribution has a mode at 2 but counts across the range. It is not clear that these distributions are reliably different. Provide confidence intervals on the modes or perform a permutation test on the difference in estimated MPC counts, and specify how many independent channel snapshots contribute to each histogram (the text mentions 100 channel estimates per measurement but not how many positions/orientations/antennas are used and whether the 100 are independent).
  3. [Section V-C] The positive/negative region analysis in Fig. 7 is used to reinforce the central claim, but the criteria for defining a P-region or N-region are not stated. Without a quantitative definition (e.g., a threshold on the difference between MUSIC-PDPs or on path gains), the selection of green and red regions appears anecdotal. Please specify the procedure and report, across all measurement locations, the frequency with which new paths appear at 6.5 GHz versus the frequency with which paths are suppressed at 8.75 GHz, rather than only one illustrative case.
minor comments (6)
  1. [Section IV-C] The sampling frequency is stated as B = 983.04 Hz, but it should presumably be 983.04 MHz given the occupied bandwidth of approximately 500 MHz; please correct the units.
  2. [Figures 4 and 5] The axis label 'Occurence' is misspelled; it should be 'Occurrence'. Also, the x-axis ranges are inconsistent across the subplots (e.g., Fig. 5 top starts at 2 while the others start at 0), which makes visual comparison harder; please unify them.
  3. [Section III-A] Please specify the frequency-smoothing subarray size and overlap used in the measurements, since these parameters affect both the model-order estimate and the MUSIC resolution.
  4. [Section II-B] After Eq. (2), the text states that the total number of delays La is the same for both frequencies, but the observed bLa differs. Please clarify explicitly that bLa is the number of distinguishable (resolvable) components, which may differ from the geometric La, and that the mode shifts in Section V-B reflect resolvability rather than physical path count.
  5. [Equation (5)] The expectation in C_{a,f_c} is not defined operationally; please specify that a sample covariance over the 100 channel estimates is used and how the smoothing is applied.
  6. [References] Reference [9] has incomplete page information ('pp. 1-1'); please update to the final published details if available.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper reports empirical measurement statistics derived with a standard estimator, and the self-citations are methodological rather than load-bearing.

full rationale

The paper's central claim is an empirical observation from measurements: histograms of the estimated number of multipath components shift upward at 6.5 GHz when a target is present and downward at 8.75 GHz. These counts are produced by a second-order elbow method applied to a frequency-smoothed covariance matrix followed by MUSIC-PDP processing; no parameter is fitted to the target/no-target labels and no equation is calibrated to reproduce the reported histogram modes. The self-citations to the Pi-Radio platform ([6]) and to prior MUSIC/channel-estimation formulations ([12], [13]) supply hardware and standard signal-processing tools; they do not assert a uniqueness theorem, impose the frequency-dependent conclusion, or smuggle in an ansatz that already contains the result. The system model's assumption that true delays and the total number of paths La are geometry-dependent and hence the same across frequencies (Section II-B) is not contradicted in a circular way because the paper consistently frames the measured quantity as 'distinguishable' multipath components, i.e., the estimator's resolvability outcome rather than the true La. A legitimate validity concern remains: the elbow estimator is not calibrated against synthetic channels, and the antenna is used at 8.75 GHz outside its specified 6.0-8.5 GHz band, so frequency-dependent SNR or eigenvalue profiles could bias bLa and produce the observed shifts as an artifact. That is a measurement-validity or robustness risk, not circularity, because the conclusion does not reduce by construction to an input, a fitted constant, or a self-citation chain. The paper is self-contained as a measurement study and its results are externally falsifiable by repeating the campaign with calibrated estimators or different hardware.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claims rest on several unstated implementation choices: the smoothing configuration and elbow threshold that produce the path counts, the MUSIC peak selection, and the antenna's unquantified operation at 8.75 GHz. The model also assumes time-invariance and frequency-independent delay structure. No new physical entities are introduced.

free parameters (3)
  • Frequency-smoothing subarray size and overlap
    Section III-A applies space-frequency smoothing to restore covariance rank; the subarray dimension and overlap determine the eigenvalue profile and hence the elbow estimate of the number of multipath components, which is the basis for the central comparison. Exact values are not reported.
  • Second-order elbow criterion parameter
    The elbow point in the sorted eigenvalues is detected via 'second-order statistics of the eigenvalues' (Section III-A), which requires a heuristic threshold. Different choices would shift the estimated path counts relative to the true number of paths.
  • MUSIC peak selection threshold and delay grid
    For each symbol the peaks of the MUSIC-PDP are collected for K-means clustering (Section III-B); the amplitude threshold and delay search grid control which paths are retained, and these settings are not specified.
assumptions (4)
  • domain assumption The channel is time-invariant during the measurement and the multipath delays are identical at 6.5 and 8.75 GHz.
    Eq. (2) models the CIR with delays tau_l,a and La independent of frequency; if the delay structure differs with frequency (for example due to frequency-dependent reflection points), the comparison of path counts across frequencies is confounded. This is assumed, not measured.
  • domain assumption The first path delays satisfy tau0,a approx tau0,b < 1/B so that MISO channel separation by time-domain windowing (Eq. 4) is valid.
    Section II-C states the first delays must be close and within one sample; this is forced by shifting the received signal, but the validity at both frequencies is assumed.
  • domain assumption Frequency-domain smoothing restores the rank of the covariance matrix so that the signal subspace dimension equals the true number of multipath components.
    Section III-A relies on spatio-frequential smoothing (after [16]) to decorrelate coherent multipath; the equivalence between the smoothed covariance rank and La is assumed, and any residual rank deficiency biases the elbow estimator.
  • ad hoc to paper The Kyocera 1005193 antenna, rated for 6.0-8.5 GHz, performs adequately at 8.75 GHz.
    Section IV-A notes 'We have experimentally verified that these antennas can be used at 8.75 GHz' without reporting gain, impedance, or pattern measurements; if the antenna is inefficient or pattern-distorted at 8.75 GHz, the multipath and blockage comparison is degraded.

how reviews work

0 comments
Cite this review

Pith. "Pith review of An Experimental Multi-Band Channel Characterization in the Upper Mid-Band." pith.science (2026). https://pith.science/paper/GYX4E7JE

@misc{pith2026241112888,
  author       = {Pith},
  title        = {Pith review of: An Experimental Multi-Band Channel Characterization in the Upper Mid-Band},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GYX4E7JE}},
  note         = {Machine review of arXiv:2411.12888}
}
read the original abstract

The following paper provides a multi-band channel measurement analysis on the frequency range (FR)3. This study focuses on the FR3 low frequencies 6.5 GHz and 8.75 GHz with a setup tailored to the context of integrated sensing and communication (ISAC), where the data are collected with and without the presence of a target. A method based on multiple signal classification (MUSIC) is used to refine the delays of the channel impulse response estimates. The results reveal that the channel at the lower frequency 6.5 GHz has additional distinguishable multipath components in the presence of the target, while the one associated with the higher frequency 8.75 GHz has more blockage. The set of results reported in this paper serves as a benchmark for future multi-band studies in the FR3 spectrum.

Figures

Figures reproduced from arXiv: 2411.12888 by the authors.

Figure 1
Figure 1. Measurement environment at NYU Wireless, Brooklyn, [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Locations and antenna orientations setup. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 4
Figure 4. Frequency 6.5 GHz. Histograms of the estimated number of multipath components, with and without target. 2 4 6 8 10 12 14 16 0 100 200 300 400 Lˆ O c c u r e n c e w/o target w/ target Orientation: Beta 0 2 4 6 8 10 12 14 16 18 20 0 50 100 150 200 250 Lˆ O c c u r e n c e w/o target w/ target Orientation: Alpha 0 2 4 6 8 10 12 14 16 18 20 0 100 200 300 400 500 600 Lˆ O c c u r e n c e w/o target w/ target Orientation… view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: Frequency 8.75 GHz. Histograms of the estimated number of multipath components, with and without target. 0 20 40 60 80 100 120 140 160 180 200 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Clustered index C l u s t e r e d e s t. dis t. 6.5 GHz 8.75 GHz [PITH_FULL_IMAGE:…
Figure 6
Figure 6. Figure 6: The K−means clustered paths for 6.5 GHz and 8.75 GHz showing path distances as a function of their index. at 3 without a target, which rises to 4 when the target reflector is introduced. Notably, at the higher frequency of 8.75 GHz, illustrated in [PITH_FULL_IMAGE:fig…
Figure 7
Figure 7. Figure 7: MUSIC-PDP with P-region (green shaded) and N-region (red shaded), with and without target (T). available, this can be an invaluable source of information to infer the target’s spatial characteristics. The results are shown in [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Near-Field Measurement System for the Upper Mid-Band

    eess.SP 2024-12 reject novelty 6.0 of 10

    A synthetic-aperture method that estimates near-field multipath parameters by triangulating reflection image points, using small non-coherent antenna arrays; validation is limited to a qualitative simulation and an un...

Reference graph

Works this paper leans on

17 extracted references · 14 canonical work pages · cited by 1 Pith paper

  1. [1]

    Twelve scientific challenges for 6G: Rethinking the foundations of communications theory,

    M. Chafii, L. Bariah, S. Muhaidat, and M. Debbah, “Twelve scientific challenges for 6G: Rethinking the foundations of communications theory,” IEEE Commun. Surv. Tutor ., vol. 25, no. 2, pp. 868–904, 2023

    work page 2023

  2. [2]

    Cellular wireless networks in the upper mid-band,

    S. Kang, M. Mezzavilla, S. Rangan, A. Madanayake, S. B. Venkatakrish- nan, G. Hellbourg, M. Ghosh, H. Rahmani, and A. Dhananjay, “Cellular wireless networks in the upper mid-band,” IEEE Open J. Commun. Soc. , Mar. 2024

    work page 2024

  3. [3]

    Notice of Proposed Rulemaking: Expanding Flexible Use of the 12.2-12.7 GHz Band,

    FCC, “Notice of Proposed Rulemaking: Expanding Flexible Use of the 12.2-12.7 GHz Band,” Docket No. 20-443 , vol. FCC-21-13, 2021

    work page 2021

  4. [4]

    Point data for site-specific mid-band radio propagation channel statistics in the indoor hotspot (InH) environment for 3GPP and next generation alliance (NGA) channel modeling,

    T. S. Rappaport, D. Shakya, and M. Ying, “Point data for site-specific mid-band radio propagation channel statistics in the indoor hotspot (InH) environment for 3GPP and next generation alliance (NGA) channel modeling,” arXiv preprint arXiv:2409.19873 , 2024

    arXiv 2024

  5. [5]

    Urban Outdoor Propagation Measurements and Channel Models at 6.75 GHz FR1(C) and 16.95 GHz FR3 Upper Mid- band Spectrum for 5G and 6G,

    D. Shakya et al. , “Urban Outdoor Propagation Measurements and Channel Models at 6.75 GHz FR1(C) and 16.95 GHz FR3 Upper Mid- band Spectrum for 5G and 6G,” in IEEE International Conference on Communications (ICC) , Montreal, Canada, June 2025, pp. 1–6, (submitted)

    work page 2025

  6. [6]

    A frequency hopping software-defined radio platform for communications and sens- ing in the upper mid-band,

    M. Mezzavilla, A. Dhananjay, M. Zappe, and S. Rangan, “A frequency hopping software-defined radio platform for communications and sens- ing in the upper mid-band,” in Proc. IEEE 25th International Workshop on Signal Processing Advances in Wireless Communications (SPA WC) , Lucca, Italy, 2024, pp. 611–615

    work page 2024

  7. [7]

    Channel mod- eling for FR3 upper mid-band via Generative Adversarial Networks,

    Y . Hu, M. Yin, M. Mezzavilla, H. Guo, and S. Rangan, “Channel mod- eling for FR3 upper mid-band via Generative Adversarial Networks,” in Proc. IEEE SPA WC, Lucca, Italy, 2024, pp. 776–780

    work page 2024

  8. [8]

    Integrated sensing and communications: Toward dual- functional wireless networks for 6G and beyond,

    F. Liu et al. , “Integrated sensing and communications: Toward dual- functional wireless networks for 6G and beyond,” IEEE J. Sel. Areas Commun., vol. 40, no. 6, pp. 1728–1767, 2022

    2022

Show all 17 references
  1. [9]

    Unique word-based frame design for bistatic integrated sensing and communication,

    R. Bomfin and M. Chafii, “Unique word-based frame design for bistatic integrated sensing and communication,” pp. 1–1, 2024

    2024

  2. [10]

    On outage-based beamforming design for dual-functional radar-communication 6G systems,

    A. Bazzi and M. Chafii, “On outage-based beamforming design for dual-functional radar-communication 6G systems,” IEEE Transactions on Wireless Communications , vol. 22, no. 8, pp. 5598–5612, 2023

    2023

  3. [11]

    Secure Full Duplex Integrated Sensing and Communications,

    ——, “Secure Full Duplex Integrated Sensing and Communications,” IEEE Transactions on Information F orensics and Security , vol. 19, pp. 2082–2097, 2024

    2024

  4. [12]

    Experimental validation of superresolution delay estimation algorithm using a 26 GHz radar setup,

    R. Bomfin, Z. Li, A. Nimr, and G. Fettweis, “Experimental validation of superresolution delay estimation algorithm using a 26 GHz radar setup,” in 2022 2nd IEEE International Symposium on Joint Communications & Sensing (JC&S) , 2022, pp. 1–6

    2022

  5. [13]

    Channel estimation for MIMO space time coded OTFS under doubly selective channels,

    R. Bomfin, M. Chafii, A. Nimr, and G. Fettweis, “Channel estimation for MIMO space time coded OTFS under doubly selective channels,” in 2021 IEEE International Conference on Communications Workshops (ICC Workshops), 2021, pp. 1–6

    2021

  6. [14]

    On mutual coupling for ULAs: Estimating AoAs in the presence of more coupling parameters,

    A. Bazzi, D. T. Slock, and L. Meilhac, “On mutual coupling for ULAs: Estimating AoAs in the presence of more coupling parameters,” in 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017, pp. 3361–3365

    2017

  7. [15]

    A statistical multipath detector for antenna array based GNSS receivers,

    P. Closas and C. Fernandez-Prades, “A statistical multipath detector for antenna array based GNSS receivers,” IEEE Transactions on Wireless Communications, vol. 10, no. 3, pp. 916–929, 2011

    2011

  8. [16]

    On spatio-frequential smoothing for joint angles and times of arrival estimation of multipaths,

    A. Bazzi, D. T. M. Slock, and L. Meilhac, “On spatio-frequential smoothing for joint angles and times of arrival estimation of multipaths,” in 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2016, pp. 3311–3315

    2016

  9. [17]

    I. T. Jolliffe, Principal component analysis for special types of data . Springer, 2002

    2002

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.