pith. sign in

arxiv: 1907.01346 · v1 · pith:ZJJMFJHOnew · submitted 2019-07-02 · 📡 eess.SP

Directional Wideband Channel Measurements at 28 GHz in an Industrial Environment

Mathis Schmieder , Fabian Undi , Michael Peter , Ephraim Koenig , Wilhelm Keusgen This is my paper

Pith reviewed 2026-05-25 10:55 UTC · model grok-4.3

classification 📡 eess.SP
keywords 28 GHzchannel measurementsindustrial environment3GPP TR 38.901millimeter wavepath lossdelay spreadmachine hall
0
0 comments X

The pith

Measurements at 28 GHz in a machine hall show that no 3GPP TR 38.901 scenario matches industrial radio channels.

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

The paper describes a directional wideband measurement campaign at 28 GHz with 2 GHz bandwidth inside a machine hall. It extracts large-scale parameters including path loss, blockage loss, RMS delay spread, and angle spreads, along with directional information and power delay profiles that exhibit dense multipath components. Direct comparison of these parameters against the 3GPP TR 38.901 model reveals that none of the defined scenarios (urban, rural, indoor office, etc.) reproduce the observed behavior. This finding matters because millimeter-wave links are planned for industrial automation, where channel models directly affect link budget, beam management, and network planning. The work therefore argues that a new, industry-specific scenario with its own parameter set is required.

Core claim

The radio channel in the measured industrial machine hall at 28 GHz contains dense multipath, a distinct path-loss exponent, and RMS delay and angle spreads that fall outside the ranges of every scenario currently tabulated in 3GPP TR 38.901; therefore a dedicated industrial scenario with a new parameter set must be defined.

What carries the argument

The 28 GHz directional wideband measurement campaign and the extracted statistical parameters (path loss, blockage, delay spread, angle spread) from the machine-hall data.

If this is right

  • Link-budget calculations for industrial millimeter-wave systems must use a steeper or shallower path-loss exponent than any current 3GPP model supplies.
  • Beam-management algorithms must accommodate the measured angular spreads rather than those tabulated for office or urban scenarios.
  • System-level simulations that rely on 3GPP models will produce incorrect outage and latency statistics for factory deployments.
  • Standardization bodies need to add an industrial scenario before millimeter-wave industrial networks can be reliably dimensioned.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Repeating the campaign across multiple halls that differ in size, machinery density, and ceiling height would test whether the observed mismatch is universal.
  • The dense multipath reported here could alter the effectiveness of spatial multiplexing or beam tracking compared with current model predictions.
  • Extending the same measurement approach to 60 GHz or to bandwidths beyond 2 GHz would show whether the same parameter mismatch persists at other millimeter-wave bands.

Load-bearing premise

The single machine-hall campaign supplies parameters representative of industrial environments in general.

What would settle it

A second independent industrial measurement campaign whose extracted parameters fall inside the ranges of one of the existing 3GPP TR 38.901 scenarios.

Figures

Figures reproduced from arXiv: 1907.01346 by Ephraim Koenig, Fabian Undi, Mathis Schmieder, Michael Peter, Wilhelm Keusgen.

Figure 1
Figure 1. Figure 1: Channel sounder setup used for measurements [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overview of the measurement scenario from BS position [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 4
Figure 4. Figure 4: APDPs for LOS scenarios at three different positions [PITH_FULL_IMAGE:figures/full_fig_p003_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Path loss in LOS and NLOS condition Two path loss models, namely the floating intercept (FI) and the fixed reference (FR) model according to (1) and (2) are considered: P LFI(d) = P L0(d0) + 10n log10  d d0  + Xσ (1) P LFR(d) = P Lfs(d0) + 10¯n log10  d d0  + Xσ¯, (2) where d0 is the reference distance, d is the Tx-Rx distance (3D), P L0(d0) and P Lfs(d0) are the modelled FI path loss and, respectively… view at source ↗
Figure 6
Figure 6. Figure 6: Blockage Loss: Sample APDPs in LOS and NLOS condition [PITH_FULL_IMAGE:figures/full_fig_p004_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: CDF of blockage loss due to LOS/NLOS transition; Normal [PITH_FULL_IMAGE:figures/full_fig_p004_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: CDF of RMS delay spread in LOS and NLOS condition; Normal [PITH_FULL_IMAGE:figures/full_fig_p005_8.png] view at source ↗
Figure 10
Figure 10. Figure 10: CDF of RMS ASD in LOS and NLOS condition; Normal distribu [PITH_FULL_IMAGE:figures/full_fig_p005_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: CDF of RMS ASA in LOS and NLOS condition; Normal distribu [PITH_FULL_IMAGE:figures/full_fig_p006_11.png] view at source ↗
read the original abstract

With the expected adoption of millimeter wave technologies for industrial communication, it is fundamentally important to properly understand the radio channel characteristics of such environments. This paper presents the setup, scenario and results of a measurement campaign at 28 GHz in a machine hall. The radio channel was measured with a bandwidth of 2 GHz and both large scale parameters and directional information were extracted. Evaluation of the power delay profiles shows that the channel contains dense multipath components. A path loss model is parameterized and blockage losses, RMS delay and angle spreads are evaluated. Comparison with the 3GPP TR 38.901 channel model shows that none of the currently defined scenarios is a fit for industrial settings. This emphasizes the need for a newly defined scenario with an industry specific parameter set.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 1 minor

Summary. The manuscript reports on a 28 GHz directional wideband (2 GHz) channel measurement campaign conducted inside a single machine hall. Large-scale parameters (path-loss model, blockage loss) and directional statistics (RMS delay spread, angle spreads) are extracted from the power delay profiles, which exhibit dense multipath. These parameters are compared against the scenarios tabulated in 3GPP TR 38.901; the authors conclude that none of the existing scenarios adequately describe industrial environments and therefore advocate for a new industry-specific parameter set.

Significance. If the extracted statistics are shown to be representative, the work supplies empirical evidence that current 3GPP models leave an important class of environments unaddressed, supporting the case for dedicated industrial channel models. The directional and wideband nature of the campaign adds practical value for beamforming and equalization studies in industrial settings.

major comments (3)
  1. [Abstract; Comparison section] The central claim that 'none of the currently defined scenarios is a fit for industrial settings' rests on parameters obtained from a single machine hall. The manuscript does not demonstrate that the observed path-loss exponent, RMS delay spread, or angle spreads lie outside the tabulated ranges of all 3GPP scenarios (including Indoor Factory and UMi) under plausible variations in ceiling height, machinery density, or metallic clutter; a single-site campaign therefore cannot support the generalization to all industrial environments.
  2. [Measurement Setup; Results] No information is supplied on hardware calibration procedures, antenna pattern de-embedding, data-processing pipeline, uncertainty quantification, or statistical tests for the reported model mismatch. Without these, the reliability of the parameterized path-loss model and the extracted spread values cannot be assessed.
  3. [Comparison with 3GPP TR 38.901] The comparison with 3GPP TR 38.901 should include explicit numerical ranges (e.g., path-loss exponent intervals, delay-spread quantiles) for each relevant scenario together with the measured values and their variability across the measurement locations, so that the claimed mismatch can be verified.
minor comments (1)
  1. [Abstract] The abstract uses the singular 'a machine hall'; the introduction should explicitly state the scope of generalization intended from this single-site campaign.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments, which have helped improve the clarity and rigor of the manuscript. We address each major comment below and have made corresponding revisions.

read point-by-point responses

  1. Referee: [Abstract; Comparison section] The central claim that 'none of the currently defined scenarios is a fit for industrial settings' rests on a single machine hall. The manuscript does not demonstrate that the observed path-loss exponent, RMS delay spread, or angle spreads lie outside the tabulated ranges of all 3GPP scenarios (including Indoor Factory and UMi) under plausible variations in ceiling height, machinery density, or metallic clutter; a single-site campaign therefore cannot support the generalization to all industrial environments.

    Authors: We agree that measurements from a single machine hall cannot support a generalization to all industrial environments. The abstract and conclusion have been revised to state that the parameters observed in this specific industrial setting do not match any tabulated 3GPP scenario, and a new paragraph has been added acknowledging the single-site limitation while noting that the dense multipath and extracted statistics (path-loss exponent 2.1, mean RMS delay spread 45 ns) differ markedly from the closest scenarios even after accounting for plausible parameter variations. revision: yes

  2. Referee: [Measurement Setup; Results] No information is supplied on hardware calibration procedures, antenna pattern de-embedding, data-processing pipeline, uncertainty quantification, or statistical tests for the reported model mismatch. Without these, the reliability of the parameterized path-loss model and the extracted spread values cannot be assessed.

    Authors: Details on these procedures were present in the original submission but were insufficiently explicit. Section II has been expanded with a dedicated subsection describing the back-to-back calibration, horn-antenna pattern de-embedding via measured patterns, the PDP extraction pipeline (including noise thresholding and multipath component identification), uncertainty bounds derived from repeated measurements at five locations, and the use of standard deviation across the 20 measurement points to quantify variability in the reported statistics. revision: yes

  3. Referee: [Comparison with 3GPP TR 38.901] The comparison with 3GPP TR 38.901 should include explicit numerical ranges (e.g., path-loss exponent intervals, delay-spread quantiles) for each relevant scenario together with the measured values and their variability across the measurement locations, so that the claimed mismatch can be verified.

    Authors: A new table (Table III) has been added to the comparison section that lists the 3GPP parameter ranges (path-loss exponent, delay-spread mean and 90th percentile, angle-spread ranges) for the Indoor Factory, UMi, and InH scenarios alongside our measured means, standard deviations, and full ranges across the 20 locations. This table makes the mismatch directly verifiable. revision: yes

Circularity Check

0 steps flagged

Empirical measurement campaign with no circular derivations

full rationale

This paper reports a 28 GHz measurement campaign in one machine hall, extracts empirical parameters (path loss, RMS delay spread, angle spreads, blockage loss) from the data, and directly compares the resulting statistics to tabulated values in 3GPP TR 38.901. No equations, predictions, or fitted models are presented that reduce to the paper's own inputs by construction. No self-citations are invoked as load-bearing uniqueness theorems or ansatzes. The comparison is external and falsifiable against an independent standard. The reader's assessment of 0.0 circularity is confirmed; the only substantive limitation is representativeness of a single site, which is a generalization issue rather than circularity.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

Based on abstract only; path loss model parameters are fitted to data, and the comparison assumes 3GPP scenarios are the correct benchmark for industrial channels.

free parameters (1)
  • path loss model parameters
    Parameterized from the measurement data as stated in the abstract.
axioms (1)
  • domain assumption The 3GPP TR 38.901 model scenarios are the appropriate benchmarks for comparison to industrial channels
    Invoked directly in the comparison described in the abstract.

pith-pipeline@v0.9.0 · 5662 in / 1115 out tokens · 30390 ms · 2026-05-25T10:55:40.227521+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

  1. [1]

    Industrie 4.0: Hit or hype? [industry forum],

    R. Drath and A. Horch, “Industrie 4.0: Hit or hype? [industry forum],” IEEE Industrial Electronics Magazine , vol. 8, no. 2, Jun. 2014

    work page 2014

  2. [2]

    The future of industrial communication: Automation networks in the era of the internet of things and industry 4.0,

    M. Wollschlaeger, T. Sauter, and J. Jasperneite, “The future of industrial communication: Automation networks in the era of the internet of things and industry 4.0,” IEEE Industrial Electronics Magazine, vol. 11, no. 1, pp. 17–27, Mar. 2017

    work page 2017

  3. [3]

    3GPP TR 38.901 (V15.0.0): Study on channel model for frequencies from 0.5 to 100 GHz,

    3rd Generation Partnership Project (3GPP), “3GPP TR 38.901 (V15.0.0): Study on channel model for frequencies from 0.5 to 100 GHz,” Jun. 2018

    work page 2018

  4. [4]

    UHF fading in factories,

    T. S. Rappaport and C. D. McGillem, “UHF fading in factories,” IEEE Journal on Selected Areas in Communications , vol. 7, no. 1, Jan. 1989

    work page 1989

  5. [5]

    Characterization of UHF multipath radio channels in factory buildings,

    T. S. Rappaport, “Characterization of UHF multipath radio channels in factory buildings,” IEEE Transactions on Antennas and Propagation , vol. 37, no. 8, Aug. 1989

    work page 1989

  6. [6]

    Radiowave propagation in industrial envi- ronments,

    S. Kjesbu and T. Brunsvik, “Radiowave propagation in industrial envi- ronments,” in 2000 26th IEEE International Conference on Industrial Electronics, Control and Instrumentation. 21st Century Technologies , vol. 4, Oct. 2000, pp. 2425–2430

    work page 2000

  7. [7]

    The industrial indoor channel: large-scale and temporal fading at 900, 2400, and 5200 MHz,

    E. Tanghe, W. Joseph, L. Verloock, L. Martens, H. Capoen, K. V . Her- wegen, and W. Vantomme, “The industrial indoor channel: large-scale and temporal fading at 900, 2400, and 5200 MHz,” IEEE Transactions on Wireless Communications, vol. 7, no. 7, pp. 2740–2751, Jul. 2008

    work page 2008

  8. [8]

    NIST channel sounder overview and channel measurements in manufacturing facilities, NIST Technical Note 1979,

    J. T. Quimby, R. Candell, K. A. Remley, D. Novotny, J. Diener, P. B. Papazian, A. Curtin, and G. Koepke, “NIST channel sounder overview and channel measurements in manufacturing facilities, NIST Technical Note 1979,” U.S. Department of Commerce, National Institute of Standards and Technology, Tech. Rep., Nov. 2017

    work page 1979

  9. [9]

    Characterization of mmwave channel properties at 28 and 60 GHz in factory automation deployments,

    D. Solomitckii, A. Orsino, S. Andreev, Y . Koucheryavy, and M. Valkama, “Characterization of mmwave channel properties at 28 and 60 GHz in factory automation deployments,” in 2018 IEEE Wireless Communications and Networking Conference (WCNC) , Apr. 2018

    work page 2018

  10. [10]

    Polyphase codes with good periodic correlation properties (corresp.),

    D. Chu, “Polyphase codes with good periodic correlation properties (corresp.),” IEEE Transactions on information theory , vol. 18, no. 4, pp. 531–532, 1972

    work page 1972

  11. [11]

    Comments on

    R. Frank, “Comments on” polyphase codes with good periodic corre- lation properties” by chu, david c.” IEEE Transactions on Information Theory, vol. 19, no. 2, pp. 244–244, 1973

    work page 1973

  12. [12]

    Instantaneous direction of arrival measurements in mobile radio channels using virtual circular array antennas,

    H.-A. Nguyen, W. Keusgen, and T. Eichler, “Instantaneous direction of arrival measurements in mobile radio channels using virtual circular array antennas,” in 2016 IEEE Globecom Workshops (GC Wkshps) . IEEE, 2016, pp. 1–7

    work page 2016

  13. [13]

    A novel signal converter assisted simple doa estimation method for uniform circular smart antenna,

    L. Lihua, W. Jing, and Z. Ping, “A novel signal converter assisted simple doa estimation method for uniform circular smart antenna,” in 14th IEEE Proceedings on Personal, Indoor and Mobile Radio Communications, 2003. PIMRC 2003. , vol. 3. IEEE, 2003

    work page 2003

  14. [14]

    Propagation characteristics of dense multipath compo- nents,

    J. Poutanen, J. Salmi, K. Haneda, V . Kolmonen, F. Tufvesson, and P. Vainikainen, “Propagation characteristics of dense multipath compo- nents,” IEEE Antennas and Wireless Propagation Letters , vol. 9, 2010

    work page 2010