Pith. sign in

REVIEW 4 major objections 6 minor 18 references

Device-Free Localization with Multiple Antenna Receivers: Simulations and Results

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a five-antenna receiver, using a diffraction-based electromagnetic body model, can estimate the angle of arrival of body-scattered radio waves and so tell which side of the radio link a passive target stands on.

desk verdict Competent but under-validated simulation study; side-discrimination claim holds only for one noiseless geometry and needs a parameter sweep before it can be taken as general. read the letter →

arxiv 2506.07784 v1 pith:F2ETIBAB submitted 2025-06-09 eess.SP

Vittorio Rampa , Federica Fieramosca , Stefano Savazzi , Michele D'Amico This is my paper
classification eess.SP
keywords device-freelocalizationelectromagneticbodymodelangleofarrivalestimationuniformlineararraybeamformingintegratedsensingandcommunicationpassiveradio
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

The paper extends a single-antenna electromagnetic body model to an antenna array and shows, by simulation, that the array's power-versus-angle response carries information about which side of the radio link a passive target occupies. If the claim holds, cheap multi-antenna WiFi-class devices could add angular information to the attenuation measurements that single-antenna device-free localization already provides. The central demonstration is a five-element uniform linear array at 2.4868 GHz: maximizing the received power ratio over angle yields an estimate that tracks the true angle for targets close to the link, breaking the left-right symmetry of single-antenna attenuation. This is a simulation study with preliminary results, not a hardware demonstration.

What carries the argument

The key machinery is an array-extended electromagnetic body model built on scalar diffraction. The target is a vertical perfectly absorbing 2-D sheet; each array element's received field is written as an integral over that sheet, and the received vector is expressed as a steering-vector term plus the body-induced perturbation. The angle-of-arrival estimator scans the planar-wave steering vector (8) and takes the angle maximizing the received power ratio (11). Because the array has multiple phase centers, the excess attenuation is no longer symmetric in the left-right direction, which is what makes side discrimination possible.

What would settle it

In the same geometry, replace the absorbing 2-D sheet with a dielectric or measured human-body model, or add a floor reflection, and check whether the arg-max of \(P_y(S=1)/P_0(S=0)\) still separates targets at \((2.5, +0.2)\) m from those at \((2.5, -0.2)\) m; if the peak no longer mirrors across the link, the side-discrimination claim fails.

Watch

Extended reading notes

Core claim

This paper claims that a uniform linear array receiver, modeled with a diffraction-based electromagnetic body model, can estimate the angle of arrival of the body-perturbed radio wave and can separate targets on the left of the link from targets on the right, at least for targets within roughly the first Fresnel zone of the link. The angle estimate is defined as the maximizer of the power ratio \(P_y(S=1)/P_0(S=0)\) over \(0 \le \gamma \le \pi\). In the simulated geometry (five antennas spaced at half a wavelength, 5 m link, target at \(x = 2.5\) m), the estimated angle \(\hat{\gamma}\) tracks the true angle for Y positions within about \(\pm 0.39\) m of the line of sight, and the excess attenuation computed at \(\hat{\gamma}\) decays for distant targets. The paper notes that the estimated angle is the direction of the distorted wave, not the geometric direction of the target, so the result is side information rather than a full position fix.

Load-bearing premise

The target is modeled as a perfectly absorbing flat sheet in a free-space room with no floor, wall, or ceiling reflections, and the estimator assumes planar wavefronts even though the simulated field is non-planar.

Editorial extensions

If this is right

  • Multi-antenna device-free receivers can jointly estimate excess attenuation and angle of arrival from the same received signals, something single-antenna links cannot do.
  • Side discrimination is usable only for targets close to the radio link, roughly within the first Fresnel zone; farther targets produce too little excess attenuation to deliver a reliable angle.
  • The model is compatible with low-cost WLAN-class multi-antenna devices, pointing toward integrated sensing and communication on WiFi6 and WiFi7 hardware.
  • The estimated angle is the direction of the body-perturbed wave rather than the target's geometric bearing, so the method supplies left-right side information rather than a complete position estimate by itself.
  • When the array has a single element the model reduces to the earlier single-antenna electromagnetic body model, so the extension preserves the existing framework as a special case.

Reading between the lines

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

  • The paper leaves implicit that fusing side-discrimination outputs from several links could triangulate a target's position: each link would contribute a left/right constraint on the target's location.
  • Using the non-planar steering vector (9) in the estimator, instead of the planar approximation (8), might widen the usable angular interval beyond the first Fresnel zone; the paper explicitly adopts the planar form as a starting hypothesis.
  • The monotone shape of the \(\hat{\gamma}\)-versus-Y curve in the near-link region suggests a calibrated mapping could interpolate a Y estimate, but the paper demonstrates only discrete left/right discrimination.
  • A direct stress test would repeat the same arg-max procedure on measured channel state information from a commercial WiFi card in a furnished room, where multipath and real body scattering replace the free-space absorbing sheet.
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

4 major / 6 minor

Summary. This paper extends a scalar diffraction electromagnetic body model for device-free localization to a uniform linear array (ULA) receiver. The received vector is written as in Eq. (7), and a beamforming-based estimator (11) is proposed to estimate the angle of arrival γ of the target-perturbed field by maximizing the normalized beamformed power over γ in [0, π]. Simulations at 2.4868 GHz with a five-antenna array and a target at x = 2.5 m show that the beamformer output peak shifts with the target's lateral position, suggesting that the array can discriminate left versus right side of the link. The paper claims this angular capability is an advantage over single-antenna DFL.

Significance. If the side-discrimination property is robust, the paper offers a simple and transparent way to add angular information to physical-statistical DFL models, with a plausible path toward WiFi6/7 CSI-based devices. The model equations (1)-(11) are clearly presented, and the proposed criterion is easy to reproduce, which are strengths. However, the central capability is currently demonstrated only in a noiseless, single-geometry simulation generated by the same model; the evidence is not yet sufficient to support the general claim. The paper is best viewed as a preliminary proof-of-concept within the authors' modeling framework.

major comments (4)
  1. [Sec. IV, Figs. 5-7] The side-discrimination claim ('the proposed method is capable of discriminating between targets placed on the left side of the link ... or right side of the link') is supported only for a single link geometry (x = 2.5 m) in a noise-free simulation. The estimator (11) is a nonlinear functional of the diffracted field; without a parameter sweep over target distance x, lateral position Y, array length M, element spacing d_a, and SNR, or an analytical condition on the objective, there is no reason to assume that sign(γ_hat - π/2) matches sign(Y) elsewhere. Please add such a sweep with noise realizations and error bars, or explicitly restrict the claim to the demonstrated configuration.
  2. [Sec. IV, Fig. 7] The horizontal axis of Fig. 7 is labeled 'true γ' but this quantity is never defined. Section III states that the estimated DoA does not coincide with the angle 'due to the location of the target.' Without an explicit definition of the ground-truth angle (e.g., the geometric angle from the array center to the target barycenter, or from the TX to the target), the plot is not interpretable, and the reader cannot distinguish a calibrated estimator from a merely monotone one. Define the true γ used in Fig. 7 and clarify its relationship to the target coordinates (x, Y).
  3. [Sec. III, Eqs. (8) and (11)] The estimator deliberately uses the planar steering vector (8) for w while the forward data are generated with non-planar propagation via (2) and (9). The paper does not quantify the effect of this mismatch on the arg-max location. Since d0 = 5 m and d_a = λ/2 are not deep far-field conditions, the quadratic phase error is small but not negligible, and the side-discrimination behavior could in principle be an artifact of the mismatch. Please report the sensitivity of γ_hat to the steering-vector assumption, for example by repeating the simulations with w from (9), or at least bound the resulting bias.
  4. [Sec. II and Sec. IV] All results are produced by the same scalar diffraction model (1) that defines the received vector; there is no comparison with full-wave simulation (e.g., [17]) or with experimental measurements. The paper should either add such independent validation or explicitly rephrase the conclusions as properties of the proposed model rather than of physical DFL channels. The current wording ('the proposed method is capable ...') overstates the evidence.
minor comments (6)
  1. [Eq. (7)] There is a stray '=' at the end of the S=1 branch: 'diag(a)E_r + n = if S=1'. Remove it.
  2. [Sec. II-A] The integration domain S is described as 'squared' but it is a rectangular sheet of dimensions 2a_y × 2a_z; use 'rectangular domain.'
  3. [Sec. IV] The Fresnel radius is reported as F_R = 0.39 m; with λ = 0.12 m and d = 5 m, the expression √(λ d/2) = √0.3 ≈ 0.547 m. The intended formula appears to be √(λ d)/2 ≈ 0.387 m; please correct the notation.
  4. [References] Reference [7] lists pages 1462-1745, which appears to be a typo; please verify the page range against the published article.
  5. [Sec. III] The sentence 'the estimated DoA ... does not coincide with the one due to the location of the target' is confusing because Fig. 7 later compares against a 'true γ.' Please explain the relationship between γ_hat and the target position.
  6. [Eq. (11)] In (11), P_y(S=1) depends on γ through the beamforming vector w, while P_0(S=0) does not; please state explicitly that w = a(γ) with a(γ) from (8) is used and that the maximization is over γ only.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the beamforming estimator is tested against an independent forward diffraction model, with no fitted parameters and no by-construction reduction.

full rationale

The paper's derivation chain is not circular. The forward model (Eqs. 1-4, 7) generates the received vector from a scalar diffraction integral over an absorbing sheet, while the estimator (Eq. 11) scans a planar-wave steering vector (Eq. 8) that is explicitly a simplification of the non-planar model (Eq. 9). The side-discrimination result shown in Figs. 5-7 is therefore a property of the model output under the estimator, not an identity built into the objective. No parameter is fitted to the target positions, and no prediction is a renamed version of an input. The self-citations to the authors' prior models [6], [7], [14], [15] are load-bearing as the source of the forward model, but the cited model has stated physical assumptions, does not include the target claim (left/right discrimination), and is externally falsifiable; this is legitimate use of prior work rather than circular import. The absence of measured data is a validation limitation, not a circularity. The paper itself acknowledges the estimated DoA 'does not coincide with the one due to the location of the target,' which further shows the comparison is not definitional. No specific equation-level reduction can be exhibited, so the appropriate finding is no significant circularity.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the scalar diffraction body model inherited from prior work, the free-space no-multipath assumption, and hand-picked simulation parameters (target size 1.8 m x 0.9 m, d0=5 m, f_c=2.4868 GHz). No new physical entity is introduced; the target is a modeling idealization.

free parameters (3)
  • Target sheet half-dimensions a_y, a_z = a_y = 0.45 m, a_z = 0.9 m
    Chosen by hand to represent an adult human (1.80 m x 0.90 m absorbing sheet). The angular discrimination interval and the magnitude of the angular response depend on this size; no sensitivity analysis is provided.
  • Central link distance d0 = d0 = 5 m
    Simulation geometry chosen by the authors; the Fresnel radius F_R = sqrt(lambda d0/2) = 0.39 m sets the interval over which the target influences the link and hence the side-discrimination range.
  • Carrier frequency f_c = 2.4868 GHz (lambda = 12 cm)
    Selected simulation carrier; wavelength determines the steering vector phases and the Fresnel radius. Results are expected to be frequency-dependent, but no frequency sweep is performed.
assumptions (5)
  • domain assumption Scalar diffraction (Kirchhoff-Fresnel) formula (1) accurately predicts the perturbed received field
    The central model is adopted from prior work [6],[14],[15] without re-derivation or experimental validation in this paper. If the scalar diffraction approximation is inaccurate for indoor body scattering, the predicted angular response is invalid.
  • domain assumption Target is a perfectly absorbing vertical 2-D sheet of height 2a_z and width 2a_y, with no transmission or reflection
    Introduced in Section II-A following [6],[7]. Real human bodies diffract and partially absorb; this idealization directly shapes the angular signature the estimator exploits.
  • domain assumption Environment is free-space with no reflections from floor, walls, or ceiling
    Stated in Section II and IV. Real indoor DFL scenarios are multipath-rich, so this is a strong simplification.
  • domain assumption Mutual antenna coupling is negligible (valid for d_a > lambda/4)
    Invoked in Section II-A before Eq. (1). For the chosen d_a = lambda/2 this is plausible, but it is not verified.
  • ad hoc to paper Planar wavefront steering vector (8) is used for beamforming and angle estimation even though the true propagation is modeled as non-planar (9)
    Section III: the approximation is assumed as a starting hypothesis because the non-planar vector requires a priori knowledge of all distances. The mismatch is acknowledged but its effect on estimation bias is not quantified.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Device-Free Localization with Multiple Antenna Receivers: Simulations and Results." pith.science (2026). https://pith.science/paper/F2ETIBAB

@misc{pith2026250607784,
  author       = {Pith},
  title        = {Pith review of: Device-Free Localization with Multiple Antenna Receivers: Simulations and Results},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/F2ETIBAB}},
  note         = {Machine review of arXiv:2506.07784}
}
read the original abstract

Device-Free Localization (DFL) is a passive radio method able to detect, estimate, and localize targets (e.g., human or other obstacles) that do not need to carry any electronic device. According to the Integrated Sensing And Communication (ISAC) paradigm, DFL networks exploit Radio Frequency (RF) devices, used for communication purposes, to evaluate also the excess attenuation due to targets moving in the monitored area, to estimate the target positions and movements. Several target models have been discussed in the literature to evaluate the target positions by exploiting the RF signals received by networked devices. Among these models, Electromagnetic (EM) body models emerged as an interesting research field for excess attenuation prediction using commercial RF devices. While these RF devices are usually single-antenna boards, the availability of low-cost multi-antenna devices e.g. those used in WLAN (Wireless Local Area Network) scenarios, allow us to exploit array-based signal processing techniques for DFL applications as well. Using an array-capable EM body model, this paper shows how to employ array-based processing to improve angular detection of targets. Unlike single-antenna devices that can provide only attenuation information, multi-antenna devices can provide both angular and attenuation estimates about the target location. To this end, simulations are presented and preliminary results are discussed. The proposed framework paves the way for a wider use of multi-antenna devices based, for instance, on WiFi6 and WiFi7 standards.

Figures

Figures reproduced from arXiv: 2506.07784 by the authors.

Figure 1
Figure 1. Single-link 3-D layout of an ULA with 2M + 1 receiving antennas. The array is deployed along a line at distance d = d0 from the TX and orthogonal to the LoS path connecting the T X with the RX0 device. Sect. V shows some preliminary conclusions and proposes future activities. II. EM BODY MODEL WITH MULTIPLE RX ANTENNAS According to the scalar diffraction framework adopted in [6], we briefly recall here the EM multi-… view at source ↗
Figure 3
Figure 3. Comparisons of the array factors for an ULA composed by [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Link layout used for the simulations: the antenna array is composed [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (2 more)
Figure 8
Figure 8. Figure 8: Excess attenuation for the different target positions located in (2.5,Y) [PITH_FULL_IMAGE:figures/full_fig_p005_8.png]
Figure 7
Figure 7. Figure 7: Estimated (γˆ) vs true (γ) values of the angle of arrival for different target positions located in (2.5,Y) with varying Y. V. CONCLUSIONS In this paper, we introduce a body model for linear antenna arrays capable of inferring both the presence of a target in the surro…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

18 extracted references · 15 canonical work pages

  1. [17]

    A. Z. Elsherbeni, et al., ”Antenna Analysis and Design using FEKO Electromagnetic Simulation Software,” The ACES Series on Computa- tional Electromagnetics and Engineering (CEME), SciTech Publishing, 2014

    work page 2014

  2. [1]

    Savazzi, et al., ”Device-free Radio Vision for assisted living: lever- aging wireless channel quality information for human sensing,” IEEE Signal Processing Magazine, vol

    S. Savazzi, et al., ”Device-free Radio Vision for assisted living: lever- aging wireless channel quality information for human sensing,” IEEE Signal Processing Magazine, vol. 33, no. 2, pp. 45–58, Mar. 2016

    work page 2016

  3. [2]

    Koutitas, ”Multiple human effects in body area networks,” IEEE Antennas and Wireless Propagation Letters, vol

    G. Koutitas, ”Multiple human effects in body area networks,” IEEE Antennas and Wireless Propagation Letters, vol. 9, pp. 938–941, 2010

    work page 2010

  4. [3]

    Shit, et al., ”Ubiquitous Localization (UbiLoc): A Survey and Tax- onomy on Device Free Localization for Smart World,” IEEE Communi- cations Surveys & Tutorials, vol

    R.C. Shit, et al., ”Ubiquitous Localization (UbiLoc): A Survey and Tax- onomy on Device Free Localization for Smart World,” IEEE Communi- cations Surveys & Tutorials, vol. 21, no. 4, pp. 3532–3564, Fourthquarter 2019

    work page 2019

  5. [4]

    Wilson, et al., ”Radio tomographic imaging with wireless networks,” IEEE Trans

    J. Wilson, et al., ”Radio tomographic imaging with wireless networks,” IEEE Trans. on Mobile Comp., vol. 9, no. 5, pp. 621–632, May 2010

    2010

  6. [5]

    Mohamed, et al., ”Physical-statistical channel model for off-body area network,” IEEE Antennas and Wireless Propagation Letters, vol

    M. Mohamed, et al., ”Physical-statistical channel model for off-body area network,” IEEE Antennas and Wireless Propagation Letters, vol. 16, pp. 1516–1519, 2017

    2017

  7. [6]

    Rampa, et al., ”EM Models for Passive Body Occupancy Inference,” IEEE Antennas and Wireless Propagation Letters, vol

    V . Rampa, et al., ”EM Models for Passive Body Occupancy Inference,” IEEE Antennas and Wireless Propagation Letters, vol. 16, pp. 2517- 2520, 2017

    work page 2017

  8. [7]

    Rampa, et al., ”Electromagnetic Models for Passive Detection and Localization of Multiple Bodies”, IEEE Transactions on Antennas and Propagation, vol

    V . Rampa, et al., ”Electromagnetic Models for Passive Detection and Localization of Multiple Bodies”, IEEE Transactions on Antennas and Propagation, vol. 70, no. 2, pp. 1462–1745, 2022

    work page 2022

Show all 18 references
  1. [8]

    Wang, et al., ”Device-free localisation with wireless networks based on compressive sensing,” IET Communications, vol

    J. Wang, et al., ”Device-free localisation with wireless networks based on compressive sensing,” IET Communications, vol. 6, no. 5, pp. 2395– 2403, Oct. 2012

    2012

  2. [9]

    Atif, et al., ”Wi-ESP—A tool for CSI-based Device-Free Wi-Fi Sensing (DFWS),” Journal of Computational Design and Engineering, vol

    M. Atif, et al., ”Wi-ESP—A tool for CSI-based Device-Free Wi-Fi Sensing (DFWS),” Journal of Computational Design and Engineering, vol. 7, no. 5, pp. 644–656, Oct. 2020

    2020

  3. [10]

    Shukri, et al., Enhancing the radio link quality of device-free local- ization system using directional antennas,” Proc

    S. Shukri, et al., Enhancing the radio link quality of device-free local- ization system using directional antennas,” Proc. of the 7th International Conference on Communications and Broadband Networking, pp. 1–5, Apr. 2019

    2019

  4. [11]

    Garcia, et al., ”POLAR: Passive object localization with IEEE 802.11 ad using phased antenna arrays,” Proc

    D. Garcia, et al., ”POLAR: Passive object localization with IEEE 802.11 ad using phased antenna arrays,” Proc. of the IEEE INFOCOM 2020- IEEE Conference on Computer Communications, pp. 1838–1847, Jul. 2020

    2020

  5. [12]

    Ruan, et al., ”Device-free indoor localization and tracking through Human-Object Interactions,” Proc

    W. Ruan, et al., ”Device-free indoor localization and tracking through Human-Object Interactions,” Proc. of the IEEE 17th International Sym- posium on A World of Wireless, Mobile and Multimedia Networks (WoWMoM’16), Coimbra, pp. 1–9, Jun. 2016

    2016

  6. [13]

    Ojeda, et al., ”Rx position effect on Device Free Indoor Localization in the 28 GHz band,” Proc

    V . Ojeda, et al., ”Rx position effect on Device Free Indoor Localization in the 28 GHz band,” Proc. of the IEEE Sensors Applications Sympo- sium (SAS’2022), Sundsvall, pp. 1–6, Aug. 2022

    2022

  7. [14]

    Rampa et al., ”Electromagnetic Models for Device-Free Radio Local- ization with Antenna Arrays,” Proc

    V . Rampa et al., ”Electromagnetic Models for Device-Free Radio Local- ization with Antenna Arrays,” Proc. of the IEEE-APS Topical Conf. on Antennas and Propagation in Wireless Communications, pp. 1–6, Sept. 2022

    2022

  8. [15]

    Rampa et al., ”An EM Body Model for Device-Free Localization with Multiple Antenna Receivers: A First Study,” Proc

    V . Rampa et al., ”An EM Body Model for Device-Free Localization with Multiple Antenna Receivers: A First Study,” Proc. of the IEEE-APS Top- ical Conf. on Antennas and Propagation in Wireless Communications, pp. 1–6, Oct. 2023

    2023

  9. [16]

    Fieramosca, et al., ”Modelling of the Floor Effects in Device-Free Radio Localization Applications”, Proc

    F. Fieramosca, et al., ”Modelling of the Floor Effects in Device-Free Radio Localization Applications”, Proc. of the 17th European Confer- ence on Antennas and Propagation (EuCAP’23), pp. 1–5, Florence, Mar. 2023

    2023

  10. [18]

    J. Benesty, J., et al., ”A Brief Overview of Conventional Beamforming,” in Array Beamforming with Linear Difference Equations, Springer Topics in Signal Processing, vol 20, pp.13–21, 2021

    2021

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.