Impact of Model Mismatch on DOA Estimation with MUSIC: Near-Field and Far-Field

Don-Roberts Emenonye; Harpreet S. Dhillon; R. Michael Buehrer

REVIEW 3 major objections 3 minor 19 references

Impact of Model Mismatch on DOA Estimation with MUSIC: Near-Field and Far-Field

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

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

Pith's one-line read When a receiver in the near-field wrongly assumes far-field propagation, MUSIC underestimates its own direction-of-arrival error, even as true accuracy degrades.

desk verdict Quantifies a real practical risk in MUSIC model mismatch, but the central numbers don't reproduce: the reported Fraunhofer distances and 80 GHz wavelength are internally inconsistent. read the letter →

arxiv 2502.05016 v1 pith:7TB3SXVU submitted 2025-02-07 eess.SP

Don-Roberts Emenonye , Harpreet S. Dhillon , R. Michael Buehrer This is my paper · ORCID

classification eess.SP

keywords MUSICdirection-of-arrivalestimationnear-fieldfar-fieldmodelmismatchFraunhoferdistancerange6Glocalization

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 asks what happens when a receiver in the near-field of a transmitter runs MUSIC, a standard direction-of-arrival estimator, using a beamforming model meant for the far-field. It reports, from simulations, that the mismatch degrades DOA accuracy, and worse, that the far-field model's own error estimate is too optimistic, so an operator would underestimate how badly the algorithm is doing. It also reports that MUSIC with the full near-field model estimates source range very accurately when the true distance is below the Fraunhofer distance, but that accuracy collapses beyond it. The approximate near-field model sits in between: better than far-field, worse than exact near-field. The reason this matters is that 6G systems with many antennas have enlarged near-field regions, so the regime assumption is no longer automatically safe.

What carries the argument

The machinery is the MUSIC (multiple signal classification) algorithm run with a beamforming matrix built from three different assumed models: the exact near-field model, an approximate near-field model (a parabolic-wave approximation of the spherical wavefront), and the far-field model (a planar wavefront). MUSIC estimates directions by scanning candidate array-response vectors and selecting those that are most nearly orthogonal to the noise subspace of the received-signal covariance matrix; the orthogonality test is only as good as the assumed model that generates those vectors. The paper's analysis of mismatch is organized around the Fraunhofer distance, $d_f = 2D^2/\lambda$, with $D$ the array diameter and $\lambda$ the wavelength, which marks where the near-field spherical wavefront begins to look planar to the array. When the receiver is inside this distance, the far-field beamforming vector is the wrong probe, and the paper's simulations show the consequences for DOA and range estimates.

What would settle it

Run the same mismatch experiment with a fully documented setup: a half-wavelength-spaced uniform linear or planar array, at least 10,000 Monte Carlo trials per distance, and a fine search grid (for example 0.01 degrees in angle and 0.01 m in range). If below the Fraunhofer distance the DOA root-mean-square error using the far-field beamformer does not exceed the error the far-field model predicts for itself, or if the near-field range error does not rise steeply beyond $2D^2/\lambda$, the paper's central claims would be contradicted.

Watch

Extended reading notes

Core claim

The paper's central claim is that feeding MUSIC a propagation model that does not match the actual propagation regime produces a specific, quantifiable performance drop rather than a generic degradation. In its simulations, when the receiver is in the near-field, the MUSIC beamformer built from the far-field planar-wave model yields large DOA error at short distances, and this error only falls to the level of the exact near-field model once the distance exceeds the Fraunhofer distance. At the same time, using the far-field beamformer against the far-field signal model gives essentially perfect DOA estimates, which is why the paper concludes that an incorrect far-field assumption underestimates the DOA estimation error: the algorithm's own error measure looks small while the true error against the actual near-field signal is large. For range, the exact near-field model estimates distance almost perfectly below the Fraunhofer distance and degrades sharply above it, while the approximate near-field model is less accurate even inside that distance and the far-field model cannot estimate range at all.

Load-bearing premise

The quantitative error curves rest on simulation details the paper does not report—the antenna array geometry, the number of Monte Carlo trials, and the angle and range search-grid spacings—so the size of the predicted error drop is not yet shown to be robust.

Editorial extensions

If this is right

Inside the near-field, an operator who assumes far-field propagation and uses the far-field MUSIC beamformer will see a small reported error while the true DOA error is large, so the far-field model cannot be trusted to certify localization accuracy.
The approximate near-field model improves on the far-field model but still loses accuracy in the near-field; it only becomes reliable for DOA at distances beyond the Fraunhofer distance.
Range estimation with the exact near-field model is essentially perfect inside the Fraunhofer distance and degrades sharply outside it, so range estimates should be treated as regime-dependent.
At distances greater than the Fraunhofer distance, the far-field received-signal model matches the near-field model's performance for DOA, confirming that the simpler model is sufficient there.
The exact near-field model used in the near-field delivers perfect DOA estimates in the simulations, matching the far-field model's performance in the far-field.

Reading between the lines

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

The 'underestimation' result implies a practical hazard the paper does not discuss: a near-field localization system that self-reports uncertainty from the far-field model will be overconfident, which matters for scenarios like autonomous navigation where a wrong angle with a small reported error can drive the wrong control action.
Because the simulation details behind the error magnitudes (array geometry, number of trials, search-grid resolution) are not reported, the qualitative direction of the results is likely robust, but the specific factor of underestimation should be treated as provisional until replicated with a documented grid.
The sharp boundary behavior around the Fraunhofer distance suggests a testable design rule: a receiver could compare the MUSIC spectrum residual under the near-field and far-field models and use the model with the smaller residual as a regime detector; the paper's data show the two models are distinguishable inside the near-field.
For 6G, the results imply that range estimation via MUSIC should carry a validity flag based on the Fraunhofer distance, since the algorithm itself does not indicate when its range output stops being reliable.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Julian—quick read of arXiv:2502.05016. The useful core is the question: what happens when MUSIC uses the wrong propagation model? The paper gives simulation curves showing that far-field processing in the near-field underestimates DOA error, and that near-field processing gives accurate range only inside the Fraunhofer distance. That is a practical issue for 6G localization, and the qualitative trends are plausible.

What's new: prior work on near/far model mismatch (refs [2] and [8]) looks at spectral efficiency or broader implications; this is a MUSIC-specific quantification with ANM and far-field beamforming matrices. It is entirely simulation-based and the paper is honest about that. No claim of new theory or algorithm.

The soft spots are real and load-bearing. The Fraunhofer distances in the figures do not follow from the stated antenna counts and frequencies. For NU=144 at 3 GHz, df=14.4 m implies an array aperture D=0.849 m, which is not described and does not match a half-wavelength ULA (which would give df≈1000 m). At 80 GHz the text says λ=3.7 cm, which is off by a factor of ten; that alone changes df by 100x. Without the array geometry, the classification of test distances as near or far is meaningless, and the claimed transition at the Fraunhofer distance is not reproducible. The paper also reports no error bars, trial counts, or search-grid spacing, so I have no idea how much the curves depend on implementation.

That said, the central qualitative claim—mismatch hurts, and assuming far-field hides the error—is almost certainly correct and consistent with the prior literature. The paper is not conceptually confused; it has a clear setup and the simulations support the broad picture. But a load-bearing numerical inconsistency like this needs to be fixed before the numbers can be trusted.

Who this is for: engineers choosing between near- and far-field processing in 6G arrays. They'd get general guidance, not exact thresholds. Does it deserve a serious referee? Yes—it is relevant and the flaws look fixable, but the review must demand the missing simulation details and corrected wavelength. If I were the editor, I'd send it to review and expect a major revision.

Referee Report

3 major / 3 minor

Summary. This paper presents a simulation study of MUSIC-based DOA and range estimation under model mismatch between the assumed propagation regime and the actual one. The authors consider four combinations: near-field received signal with near-field, approximate near-field (ANM), and far-field beamforming models, plus the matched far-field case. Simulations are run for fc=3GHz and fc=80GHz, with NU=144 or 256 antennas, SNR=30dB (10dB for the range figure), and source distances from 0.2m to 30m. The main claims are that using the far-field model in the near-field causes a performance drop, that incorrectly assuming far-field propagation leads to an underestimation of the DOA estimation error, and that near-field MUSIC gives accurate range estimates inside the Fraunhofer distance but degrades beyond it.

Significance. If the results are robust, this is a useful cautionary study for 6G localization: it demonstrates that model mismatch in MUSIC is not benign, and it identifies the Fraunhofer distance as a meaningful boundary for range estimation. A clear strength is that the paper contains no fitted parameters; the plotted curves are direct outputs of the MUSIC algorithm on synthetic data, so the qualitative trends are not obtained by curve fitting. However, the quantitative claims are currently limited by missing reproducibility details (array geometry, search-grid resolution, Monte Carlo trial counts), and the stated Fraunhofer distances cannot be reconstructed from the parameters given. The work is therefore a promising and relevant contribution, but it needs additional reporting before the quantitative conclusions can be accepted.

major comments (3)

[Section IV, Figs. 1-4] The Fraunhofer distances reported in the paper are not reproducible from the stated parameters (NU and fc) because the array aperture D is never specified. With df = 2D^2/λ, the value df = 14.4 m for NU=144 at fc=3GHz implies D ≈ 0.849 m, whereas a conventional half-wavelength ULA with 144 elements would have D = 7.15 m and df ≈ 1022 m, placing all test distances (0.2-30 m) in the near-field. Since the classification of test distances as inside or outside the near-field drives the central claims of a transition at df and of an 'underestimated' DOA error, the array geometry must be stated explicitly so that the simulation setup can be reproduced and the claimed boundary verified.
[Section IV, Figs. 1-5] The simulation results are presented as single curves with no error bars, no stated number of Monte Carlo trials, and no sensitivity analysis over array geometry or MUSIC search-grid resolution. The abstract and conclusion say that the loss in performance is 'quantified,' but with only one deterministic-looking trace per configuration, the numerical magnitudes — including the amount of error underestimation — are not yet supported. Please report the number of trials, the grid spacings for azimuth, elevation, and range, and at least one error-bar or percentile representation so that the quantitative strength of the claims can be assessed.
[Section IV, Figs. 3-4] The text states that the operating wavelength is 3.7 cm for fc = 80 GHz, but fc = 80 GHz corresponds to λ = 3.75 mm, and the reported Fraunhofer distance df = 0.54 m for NU = 144 is consistent with the millimetre wavelength, not with 3.7 cm. This inconsistency must be corrected because the reader cannot otherwise verify the Fraunhofer-distance calculations that underlie the near-field/far-field classification in those figures.

minor comments (3)

[Section II, Eq. (2)] In the definition of the far-field array response, the subscript Δ_ku should be Δ_kU for consistency with the rest of the equation; the current notation mixes the element index u with the destination index U.
[Section II, Fraunhofer distance definition] The phrase 'maximum diameter among the source and destination surface diameters' is imprecise for an antenna array. The diameter D in df = 2D^2/λ should be defined explicitly, for example as the maximum distance between any two antenna elements of the receive array.
[Section IV, Fig. 5] The range-estimation figure would be more informative if the paper stated the range search grid spacing and the number of candidate range values used in the near-field MUSIC search, since the accuracy of the range estimate can depend on that discretization.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the reported curves are direct MUSIC simulation outputs with no fitted parameters or self-referential predictions; the missing antenna geometry is a reproducibility gap, not a circular step.

full rationale

This paper is a forward simulation study: every claimed result is the direct output of the MUSIC null-spectrum search run on independently generated noisy snapshots, with no parameter fitted to a subset of the data and then repackaged as a prediction. The 'underestimation' claim (abstract; reiterated in Section V) compares two simulated scenarios, near-field data processed with the far-field beamformer versus far-field data processed with the same beamformer, and the reported difference is an algorithmic consequence of the model mismatch, not an input to the simulation. The range-accuracy claim (Section IV, Fig. 5) likewise emerges from the identifiability of the range parameter in the near-field steering vectors; the Fraunhofer boundary df = 2D^2/λ is a standard external definition, and although it is cited to the authors' prior work [5], the formula does not depend on the present paper's fitted values, so the self-citation is not load-bearing. The paper contains no ansatz smuggled in via citation and no imported uniqueness theorem. The genuine weakness is reproducibility rather than circularity: Section IV reports NU and fc but never the array geometry or aperture D, so the stated Fraunhofer distances (df = 14.4 m, 25.6 m, 0.54 m, 0.96 m) cannot be independently verified; the 80 GHz case also states λ = 3.7 cm, which is inconsistent with fc = 80 GHz (λ ≈ 3.75 mm). Under the analysis rules, missing experimental detail is a correctness-risk concern and does not raise the circularity score, which remains 0.

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

The paper introduces no fitted parameters or invented entities. It depends on standard MUSIC assumptions, the near/far-field channel models, and an implicit choice of array geometry and search grid. The quantitative results are simulation outputs, not derivations.

assumptions (3)

domain assumption MUSIC subspace orthogonality holds: signal and noise subspaces are uncorrelated and the sample covariance approximates the true covariance.
Section III relies on this to construct the MUSIC spectrum and to justify searching over beamforming vectors.
domain assumption Far-field plane wave approximation and Fraunhofer distance boundary define the mismatch scenarios.
The mismatch study uses the far-field model as the incorrect model in the near-field and uses the Fraunhofer distance as the transition boundary.
domain assumption Source signals are independent, so the transmitted covariance matrix RX is diagonal.
Used in Section III to separate signal and noise subspaces and to justify the MUSIC orthogonality condition.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Impact of Model Mismatch on DOA Estimation with MUSIC: Near-Field and Far-Field." pith.science (2026). https://pith.science/paper/7TB3SXVU

@misc{pith2026250205016,
  author       = {Pith},
  title        = {Pith review of: Impact of Model Mismatch on DOA Estimation with MUSIC: Near-Field and Far-Field},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7TB3SXVU}},
  note         = {Machine review of arXiv:2502.05016}
}

read the original abstract

There has been substantial work on developing variants of the multiple signal classification (MUSIC) algorithms that take advantage of the information present in the near-field propagation regime. However, it is not always easy to determine the correct propagation regime, which opens the possibility of incorrectly applying simpler algorithms (meant for far-field) in the near-field regime. Inspired by this, we use simulation results to investigate the performance drop when there is a mismatch between the signal model in the MUSIC algorithm and the propagation regime. For direction of arrival (DOA) estimation, we consider the cases when the receiver is in the near-field region but uses i) the near-field model, ii) the approximate near-field model (ANM) model, and iii) the far-field model to design the beamforming matrix in the MUSIC algorithm. We also consider the case when the receiver is in the far-field region, and we use the correct far-field model to design the beamforming matrix in the MUSIC algorithm. One contribution is that in the near-field, we have quantified the loss in performance when the ANM and the far-field model are used to create the beamforming matrix for the MUSIC algorithm, causing a reduction in estimation accuracy compared to the case when the correct near-field model is used to design the beamforming matrix. Another result is that in the near-field, when we incorrectly assume that the receiver is in the far-field and subsequently use the far-field beamforming matrix, we underestimate the DOA estimation error. Finally, we show that the MUSIC algorithm can provide very accurate range estimates for distances less than the Fraunhofer distance. This estimate gradually becomes inaccurate as the distances exceed the Fraunhofer distance.

Figures

Figures reproduced from arXiv: 2502.05016 by the authors.

**Figure 4.** Error in estimating azimuth and elevation angles w [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗

**Figure 5.** Error in estimating the relative distance between [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 19 canonical work pages

[2]

Near and Far Field Model Mismatch: Implications on 6G Communications, Localization, and Sensing

A. Elzanaty, J. Liu, A. Guerra, F. Guidi, Y . Ma, and R. Tafa zolli, “Near and far ﬁeld model mismatch: Implications on 6G communicati ons, localization, and sensing,” arXiv:2310.06604, 2023

work page Pith review arXiv 2023
[8]

Spherical w avefronts improve MU-MIMO spectral efﬁciency when using electricall y large arrays,

G. Bacci, L. Sanguinetti, and E. Bj¨ ornson, “Spherical w avefronts improve MU-MIMO spectral efﬁciency when using electricall y large arrays,” IEEE Wireless Commun. Letters , vol. 12, no. 7, pp. 1219–1223, Apr. 2023

work page 2023
[1]

Localization of signals in the near-ﬁe ld of an antenna array,

B. Friedlander, “Localization of signals in the near-ﬁe ld of an antenna array,” IEEE Trans. on Signal Processing , vol. 67, no. 15, pp. 3885– 3893, Jun. 2019

work page 2019
[3]

Performance analysis of hig her order esprit for localization of near-ﬁeld sources,

N. Y uen and B. Friedlander, “Performance analysis of hig her order esprit for localization of near-ﬁeld sources,” IEEE Trans. on Signal Processing, vol. 46, no. 3, pp. 709–719, Mar. 1998

work page 1998
[4]

Fundam entals of RIS-aided localization in the far-ﬁeld,

D.-R. Emenonye, H. S. Dhillon, and R. M. Buehrer, “Fundam entals of RIS-aided localization in the far-ﬁeld,” accepted to IEEE Trans. on Wireless Commun., available online: arxiv.org/abs/2206. 01652, 2022

work page 2022
[5]

RIS-Aided Localization under Position and Orientation Offsets in the Near and Far Field

——, “RIS-aided localization under position and orienta tion offsets in the near and far ﬁeld,” accepted to IEEE Trans. on Wireless Commun., available online: arxiv.org/abs/2210.03599 , 2022

work page Pith review arXiv 2022
[6]

Estimation of RIS misorientation in both near and fa r ﬁeld regimes,

——, “Estimation of RIS misorientation in both near and fa r ﬁeld regimes,” accepted to Proc., IEEE Intl. Conf. on Commun. (ICC) , 2023

work page 2023
[7]

Limitations of RIS-aided localization: Inspectin g the relation- ships between channel parameters,

——, “Limitations of RIS-aided localization: Inspectin g the relation- ships between channel parameters,” accepted to Proc., IEEE Intl. Conf. on Commun. (ICC) W orkshop , 2023

work page 2023

Show all 19 references

[9]

Localiza tion of near-ﬁeld sources: A reduced-dimension music algorithm ,

X. Zhang, W. Chen, W. Zheng, Z. Xia, and Y . Wang, “Localiza tion of near-ﬁeld sources: A reduced-dimension music algorithm ,” IEEE Commun. Letters , vol. 22, no. 7, pp. 1422–1425, July. 2018

2018
[10]

Joint DOA, range, and polar ization estimation for rectilinear sources with a cold array,

H. Chen, W. Wang, and W. Liu, “Joint DOA, range, and polar ization estimation for rectilinear sources with a cold array,” IEEE Wireless Commun. Letters , vol. 8, no. 5, pp. 1398–1401, Oct. 2019

2019
[11]

J. Xu, B. Wang, and F. Hu, IEEE Wireless Commun. Letters , vol. 8, no. 1, pp. 89–92, Feb. 2019

2019
[12]

Simpliﬁed high-order DOA and range estimation with linear antenna array,

J. Li, Y . Wang, C. Le Bastard, G. Wei, B. Ma, M. Sun, and Z. Y u, “Simpliﬁed high-order DOA and range estimation with linear antenna array,” IEEE Commun. Letters , vol. 21, no. 1, pp. 76–79, Jan. 2017

2017
[13]

Subspac e-based algorithms for localization and tracking of multiple near- ﬁeld sources,

W. Zuo, J. Xin, H. Ohmori, N. Zheng, and A. Sano, “Subspac e-based algorithms for localization and tracking of multiple near- ﬁeld sources,” IEEE J. of Sel. Topics in Signal Processing , vol. 13, no. 1, pp. 156–171, Mar. 2019

2019
[14]

Subspac e-based near-ﬁeld source localization in unknown spatially nonuni form noise environment,

W. Zuo, J. Xin, N. Zheng, H. Ohmori, and A. Sano, “Subspac e-based near-ﬁeld source localization in unknown spatially nonuni form noise environment,” IEEE Trans. on Signal Processing , vol. 68, pp. 4713– 4726, Aug. 2020

2020
[15]

Mixed-order music algorit hm for localization of far-ﬁeld and near-ﬁeld sources,

B. Wang, Y . Zhao, and J. Liu, “Mixed-order music algorit hm for localization of far-ﬁeld and near-ﬁeld sources,” IEEE Signal Processing Letters, vol. 20, no. 4, pp. 311–314, Apr. 2013

2013
[16]

Passive localization of mixed near -ﬁeld and far- ﬁeld sources using two-stage music algorithm,

J. Liang and D. Liu, “Passive localization of mixed near -ﬁeld and far- ﬁeld sources using two-stage music algorithm,” IEEE Trans. on Signal Processing, vol. 58, no. 1, pp. 108–120, Jan 2010

2010
[17]

Phase compensation -based localization of mixed far-ﬁeld and near-ﬁeld sources,

Y . Tian, X. Gao, W. Liu, and H. Chen, “Phase compensation -based localization of mixed far-ﬁeld and near-ﬁeld sources,” IEEE Wireless Commun. Letters , vol. 11, no. 3, pp. 598–601, Mar. 2022

2022
[18]

Mixed near -ﬁeld and far-ﬁeld source localization based on convolution neural n etworks via symmetric nested array,

X. Su, P . Hu, Z. Liu, T. Liu, B. Peng, and X. Li, “Mixed near -ﬁeld and far-ﬁeld source localization based on convolution neural n etworks via symmetric nested array,” IEEE Trans. V eh. Technol., vol. 70, no. 8, pp. 7908–7920, Aug. 2021

2021
[19]

Complex resne t aided DOA estimation for near-ﬁeld MIMO systems,

Y . Cao, T. Lv, Z. Lin, P . Huang, and F. Lin, “Complex resne t aided DOA estimation for near-ﬁeld MIMO systems,” IEEE Trans. V eh. Technol. , vol. 69, no. 10, pp. 11 139–11 151, Oct. 2020

2020

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [2]

Near and Far Field Model Mismatch: Implications on 6G Communications, Localization, and Sensing

A. Elzanaty, J. Liu, A. Guerra, F. Guidi, Y . Ma, and R. Tafa zolli, “Near and far ﬁeld model mismatch: Implications on 6G communicati ons, localization, and sensing,” arXiv:2310.06604, 2023

work page Pith review arXiv 2023

[2] [8]

Spherical w avefronts improve MU-MIMO spectral efﬁciency when using electricall y large arrays,

G. Bacci, L. Sanguinetti, and E. Bj¨ ornson, “Spherical w avefronts improve MU-MIMO spectral efﬁciency when using electricall y large arrays,” IEEE Wireless Commun. Letters , vol. 12, no. 7, pp. 1219–1223, Apr. 2023

work page 2023

[3] [1]

Localization of signals in the near-ﬁe ld of an antenna array,

B. Friedlander, “Localization of signals in the near-ﬁe ld of an antenna array,” IEEE Trans. on Signal Processing , vol. 67, no. 15, pp. 3885– 3893, Jun. 2019

work page 2019

[4] [3]

Performance analysis of hig her order esprit for localization of near-ﬁeld sources,

N. Y uen and B. Friedlander, “Performance analysis of hig her order esprit for localization of near-ﬁeld sources,” IEEE Trans. on Signal Processing, vol. 46, no. 3, pp. 709–719, Mar. 1998

work page 1998

[5] [4]

Fundam entals of RIS-aided localization in the far-ﬁeld,

D.-R. Emenonye, H. S. Dhillon, and R. M. Buehrer, “Fundam entals of RIS-aided localization in the far-ﬁeld,” accepted to IEEE Trans. on Wireless Commun., available online: arxiv.org/abs/2206. 01652, 2022

work page 2022

[6] [5]

RIS-Aided Localization under Position and Orientation Offsets in the Near and Far Field

——, “RIS-aided localization under position and orienta tion offsets in the near and far ﬁeld,” accepted to IEEE Trans. on Wireless Commun., available online: arxiv.org/abs/2210.03599 , 2022

work page Pith review arXiv 2022

[7] [6]

Estimation of RIS misorientation in both near and fa r ﬁeld regimes,

——, “Estimation of RIS misorientation in both near and fa r ﬁeld regimes,” accepted to Proc., IEEE Intl. Conf. on Commun. (ICC) , 2023

work page 2023

[8] [7]

Limitations of RIS-aided localization: Inspectin g the relation- ships between channel parameters,

——, “Limitations of RIS-aided localization: Inspectin g the relation- ships between channel parameters,” accepted to Proc., IEEE Intl. Conf. on Commun. (ICC) W orkshop , 2023

work page 2023