A Grant-free Coded Random Access Scheme for Near-field Communications

Davide Dardari; Enrico Paolini; Enrico Testi; Giulia Torcolacci; Nicol\`o Decarli

REVIEW 5 major objections 4 minor 20 references

A Grant-free Coded Random Access Scheme for Near-field Communications

T0 review · 5 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Near-field spatial modes can carry grant-free access without channel estimation.

desk verdict A solid incremental CRA scheme for near-field massive access; the core premise holds for their parameters, but fix power normalization and position assumption. read the letter →

arxiv 2508.15673 v1 pith:QEDI2RYC submitted 2025-08-21 eess.SP

Enrico Testi , Giulia Torcolacci , Nicolò Decarli , Davide Dardari , Enrico Paolini This is my paper · ORCID

classification eess.SP

keywords grant-freerandomaccesscodednear-fieldcommunicationsextremelylargeaperturearraysuccessiveinterferencecancellationmassivemachine-typespatialmultiplexingindustrialIoT

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 sets out to show that the parallel spatial channels created when an extremely large aperture array operates in the near field can be used as extra replica resources in a coded random access protocol. It proposes a grant-free scheme, CSRA, in which each active machine transmits copies of its packet along several beam directions, and the access point clusters antenna elements that see the same pilot, combines them, and runs successive interference cancellation. The simulations indicate that this design reaches packet loss rates around 10^-4 with light load and a few times 10^-3 with 45 active users, while processing each antenna element independently cannot reach 10^-2 even with five users. The reason to care is that dense, sporadic industrial traffic could then be handled without scheduling or channel estimation, using only the geometry of the array.

What carries the argument

The central object is the set of near-field communication modes: parallel LOS channels created by spherical wavefronts between a small user array and a much larger aperture array. Eq. (3) counts them as floor(1 + 2 L_T L_R / (lambda sqrt(4 d^2 + L_R^2))). The paper's mechanism is to realize these modes with DFT beam-steering vectors, so that a user can transmit independent replicas in distinct directions; the AP then clusters contiguous antenna elements per pilot, applies MRC inside each cluster to get array gain, and runs SIC to resolve collisions.

What would settle it

Take a 20-element transmitting ULA and a 20 m, 60 GHz ELAA separated by a distance inside the Fresnel region, and measure the singular-value spectrum of the LOS channel matrix. If fewer than floor(1 + 2 L_T L_R / (lambda sqrt(4 d^2 + L_R^2))) singular values are above the coupling threshold, or if the DFT beams of Eq. (6) have pairwise correlation well above zero, then the independent-replica assumption underlying the simulated packet loss rates is violated.

Watch

Extended reading notes

Core claim

The paper proposes coded spatial random access (CSRA), a grant-free protocol in which each active network user sends R copies of its packet along R of the near-field communication modes available between its small array and the extremely large aperture array at the access point. Because the wavefront is spherical, the channel matrix has rank larger than one even in pure line-of-sight, and the paper relies on the known result that the number of strongly coupled modes is given by Eq. (3). The AP does not estimate the full channel; it detects pilots per antenna element, clusters contiguous elements that receive the same pilot, combines each cluster with MRC, decodes, and then cancels decoded re

Load-bearing premise

The load-bearing premise is that the near-field communication modes counted by Eq. (3) are actually realized by the simple DFT beamforming vectors of Eq. (6), and that those modes are orthogonal and strongly coupled enough to behave as independent replica channels.

Editorial extensions

If this is right

Dense, uncoordinated IIoT access can operate grant-free in the near field without per-user channel estimation, as long as the ELAA clusters and combines received signals.
SIC is essential: without it the simulated scheme cannot reach a packet loss rate of 10^-2, so the reliability gain is tied to iterative interference cancellation.
The optimal number of replicas R depends on load: R=4 gives near 10^-4 PLR at K=25, while R=4 or R=5 is better at K=45, suggesting an adaptive replica-count rule.
Clustering and combining at the ELAA is what enables low-PLR operation; single-element processing stays above 10^-2 even for K=5 active users.
The protocol requires fixed user positions to select the beam directions that intercept the ELAA, which fits stationary factory machinery.

Reading between the lines

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

If the mode-count premise holds, the same spatial-replica trick should work wherever the Fresnel condition holds, so the scheme's core idea is portable to other frequency bands and array sizes by rescaling Eq. (3).
Because the paper simulates only a fixed geometry, a natural extension is to let users move or tilt, which would change which DFT beams intercept the ELAA and test whether the cluster-to-beam mapping survives.
The slot-synchronous beacon model is a simplifying assumption; an asynchronous or misaligned variant would show how much of the gain depends on the common time reference.
Clusters themselves encode the user's angular position, so a future scheme could exploit cluster location for user identification or localization without extra overhead.

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

Short version: this is a legitimate new combination—using near-field spatial directions as replicas in coded random access, with clustering and SIC—and the Monte Carlo results look credible. The load-bearing premise, that DFT beams from the NU yield orthogonal effective channel vectors across the ELAA, is not actually verified in the paper, but for their specific parameters it appears to hold: NT=20, LR=4000λ means adjacent DFT beams differ by 0.1 in sine space, which is exactly an integer number of cycles over the 8000-element array. So the stress-test candidate concern about non-orthogonality does not land as fatal, though the authors should state it.

What is actually new: the idea of treating near-field spatial beams as CRA replica resources, with cluster-based MRC and SIC at the AP, sidestepping channel estimation. The protocol is fully described, and the simulations show a large improvement over single-element processing and over the no-SIC variant. That is worth having.

Soft spots, in rough order of importance. First, a power inconsistency: Eq (1) defines b(k)=1/R sum b(k,j), implying total transmit power PT, but Eqs (4)-(5) drop that normalisation, implying each replica is sent at full power PT. The paper never says which one the simulations used. If it is the latter, the absolute PLR values are optimistic; the relative comparisons may survive, but the model needs to be made consistent. Second, position knowledge is quietly assumed: the SIC step relies on knowing where each NU is to know which antenna elements its replicas hit. That should be in the system model, not implied later. Third, the latency claim in the abstract is not supported by any metric. Fourth, there is no comparison with a conventional time-domain CRA baseline; the paper only compares against its own variants. That is acceptable for a short paper, but it limits what you can say about the scheme's absolute merit.

Overall, I trust the qualitative result. The paper is not a breakthrough, but it is a plausible incremental contribution with a clear story. I would send it to peer review: a competent reviewer could get a clean revision. Worth citing if you work on CRA or near-field massive access.

Referee Report

5 major / 4 minor

Summary. The paper proposes a grant-free coded random access scheme for near-field communications, termed coded spatial random access (CSRA). Multi-antenna network users transmit multiple spatial replicas of the same packet toward an extremely large aperture array (ELAA) access point, using DFT-based beam steering directions. The AP clusters antenna elements by energy detection per pilot, performs per-cluster MRC decoding, and then runs successive interference cancellation (SIC). The authors argue that near-field communication modes provide multiple orthogonal spatial channels, enabling CRA diversity without channel estimation. Numerical Monte Carlo simulations report packet loss rate (PLR) as a function of the number of active users K and the number of replicas R, comparing CSRA with and without SIC and against a single-element processing variant (CSRA-SE). The claimed contribution is improved reliability and reduced latency for dense uncoordinated IIoT networks.

Significance. The topic is timely, and the protocol architecture is clearly described at an algorithmic level. If the central premise holds—namely, that the R spatial replicas are near-orthogonal, strongly coupled effective channels that act as independent CRA resources—the scheme could be a useful grant-free option for near-field XL-MIMO IIoT. The manuscript gives credit for not fitting any parameter to the PLR curves; the mode-count formula is taken from prior work [18] and is not extracted from the current simulations, so the qualitative trends are not circular. The simulation results are internally plausible (e.g., the existence of an optimal R depending on load). However, the current evidence does not yet establish the main claim: the power normalization is inconsistent, the mode-realization premise is not verified for the simulated geometry, and the chosen baseline does not isolate the benefit of spatial CRA relative to conventional time-domain CRA. The significance of the contribution therefore remains conditional on these points being addressed.

major comments (5)

[§III.A, Eqs. (1), (4)–(5)] There is a power-normalization inconsistency. Eq. (1) defines b^(k) = (1/R^(k)) Σ_j b^(k,j), so for unit-norm mutually orthogonal DFT beams the total radiated power per NU is PT/R^(k). Eqs. (4) and (5), however, use Σ_j b^(k,j) without the 1/R factor, corresponding to total power R^(k)·PT. The two conventions differ by a factor R^2 in power. Since the PLR curves in Sec. IV depend on SNR, the absolute PLR values may be optimistic. Please state the intended transmit-power constraint and re-run the simulations under a single consistent normalization.
[§III.B, Eq. (6)] The load-bearing premise is that the R^(k)_max modes counted by Eq. (3), taken from [18], are realized by the DFT beamformers (6), yielding near-orthogonal and strongly coupled effective channel vectors H^(k)b^(k,j). This is asserted based on [18] but not verified for the geometry used in Sec. IV (LR=20 m, d≈8 m, λ=5 mm, NT=20). The DFT codebook orthogonality in (6) is a far-field property of the NU array; it does not automatically imply orthogonality of the cascade H^(k)b^(k,j) across the finite ELAA. Please report the Gram matrix or singular-value spectrum of the effective channel for representative NU positions, and confirm that the values of R used in the simulations (2–7) correspond to near-orthogonal, strongly coupled replicas. Without this, the interpretation of the R-replica PLR gains as spatial CRA diversity is not supported.
[§IV.C] The only performance baseline is CSRA-SE, which removes clustering and combining. This comparison conflates the benefit of the proposed spatial-replica CRA with the receive-array gain obtained from cluster MRC. To support the claim that integrating near-field spatial multiplexing with CRA improves reliability, the paper should compare with a time-domain CRA scheme using the same number of replicas and the same total energy per message (e.g., R replicas in R slots with a single beam), or at least with a single-beam, non-spatial CSRA variant. The current results show that array combining helps, but they do not demonstrate that the spatial-mode diversity mechanism itself is beneficial relative to conventional CRA.
[§III.C, §IV.A] Pilot collisions are not addressed. The protocol lets each active NU draw pilots from a set of size P with P ≪ K, yet the energy detection and clustering in Eqs. (8)–(9) operate per pilot. If two active NUs choose the same pilot and their footprints overlap on the ELAA, the channel estimate in (8) is a superposition, and the cluster is ambiguous. The manuscript does not state the value of P used in the simulations, how same-pilot collisions are resolved before SIC, or whether the reported PLR already includes such collisions. This is a load-bearing detail for the dense, uncoordinated scenario claimed.
[§IV.A, Eq. (3)] For the stated geometry (λ=5 mm, NT=20, LT≈0.05 m, LR=20 m, d between 8 and 8.54 m), Eq. (3) gives Rmax ≈ 15–16, not 20. The text states 'each NU has up to Rmax = 20 orthogonal transmission directions ... leaving each NU with Rmax = 13 usable orthogonal directions.' Please reconcile the mode-count formula with the simulated geometry and report the actual distribution of usable R per NU. Since only R ≤ 7 is simulated, this does not invalidate the PLR curves, but the description of the near-field diversity is internally inconsistent.

minor comments (4)

[§III.C.1, Eq. (9)] The threshold η = 2σ/NP appears dimensionally inconsistent: |\hat h_r(pj)|² is a squared amplitude, while σ/NP has amplitude units. Please correct the expression (probably η = 2σ²/NP or similar) and include a sensitivity analysis with respect to η, since the clustering result depends on this heuristic threshold.
[§III.C.3] The SIC procedure relies on the AP knowing the NU position to identify the set of antenna elements UW. This is an important assumption; it should be stated explicitly in the system model and discussed with respect to positioning accuracy requirements.
[§II, §III.A] The assumption of perfect slot synchronization and ideal pilot orthogonality is stated, but its practical impact on the PLR results is not discussed. A sentence on synchronization error tolerance would improve the paper's applicability.
[§IV.A] The Monte Carlo simulation parameters do not include the pilot set size P, the number of Monte Carlo runs, or confidence intervals. These should be reported to make the numerical results reproducible.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the PLR results are produced by independent Monte Carlo simulation and no fitted parameter is disguised as a prediction.

full rationale

The paper's claimed contribution—that near-field spatial modes combined with coded random access and SIC at an ELAA reduce packet loss—is evaluated through direct simulation of the physical channel model in Eq. (2). The mode-count formula Eq. (3) and the DFT beam directions Eq. (6) are taken from the authors' prior work [18], but that prior result is parameter-free and geometrically determined (LT, LR, λ, distance); it is not fitted to the current PLR curves. The simulation then computes PLR from actual channel realizations and beamforming, so the performance comparison between CSRA and CSRA-SE is not forced by construction. The only self-citation of concern is [18], which supplies the number of usable communication modes; however, it is used as an independent analytical input and is not equivalent to the output PLR. A separate consistency issue exists between Eq. (1)'s 1/R normalization and Eqs. (4)-(5)'s unnormalized replica sum, but this affects power accounting and correctness, not circularity. Therefore the central derivation chain is self-contained with respect to the claimed prediction.

Assumptions & free parameters 1 free parameters · 6 assumptions · 0 invented entities

The central claim rests on the cited near-field mode theory, the free-space LOS channel model, and several position and synchronization assumptions. No new physical entities are introduced.

free parameters (1)

Clustering threshold η = η = 2σ/NP
Ad hoc detection threshold; the paper states it ensures robust initial signal detection but gives no sensitivity analysis or optimization. The packet loss results depend on this choice.

assumptions (6)

domain assumption The number of well-coupled near-field modes Rmax is given by Eq. (3) from [18].
The protocol's replica capacity rests on this cited mode-count result from the authors' prior work, which is not re-derived or measured here.
domain assumption The free-space LOS channel model (Eq. 2) with spherical wavefronts holds; no multipath or blockage.
The channel matrix is modeled as line-of-sight only, which determines the clustering and combining behavior in the simulations.
domain assumption DFT beam steering vectors (Eq. 6) realize the Rmax orthogonal modes with significant coupling.
Asserted in Sec. III.B based on [18]; if these beams are not orthogonal at the ELAA, the independent-replica premise fails.
domain assumption The AP knows the position of each NU, allowing it to identify antenna elements for SIC.
Sec. III.C.3 uses 'knowing the position of the NU' to perform interference cancellation; no position uncertainty is modeled.
domain assumption Slot synchronization is maintained by a beacon signal from the AP.
The protocol assumes a common time reference; synchronization error is not analyzed.
domain assumption Clusters of contiguous antenna elements with energy above threshold correspond to distinct NUs and can be spatially separated.
The clustering step assumes users' beams illuminate disjoint or separable ELAA regions; collisions in the same region with the same pilot are not analyzed.

how reviews work

0 comments

Cite this review

Pith. "Pith review of A Grant-free Coded Random Access Scheme for Near-field Communications." pith.science (2026). https://pith.science/paper/QEDI2RYC

@misc{pith2026250815673,
  author       = {Pith},
  title        = {Pith review of: A Grant-free Coded Random Access Scheme for Near-field Communications},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QEDI2RYC}},
  note         = {Machine review of arXiv:2508.15673}
}

read the original abstract

The industrial Internet of things (IIoT) is revolutionizing industrial processes by facilitating massive machine-type communications among countless interconnected devices. To efficiently handle the resulting large-scale and sporadic traffic, grant-free random access protocols-especially coded random access (CRA)-have emerged as scalable and reliable solutions. At the same time, advancements in wireless hardware, including extremely large-scale MIMO arrays and high-frequency communication (e.g., mmWave, Terahertz), are pushing network operations into the near-field propagation regime, allowing for dense connectivity and enhanced spatial multiplexing. This paper proposes an innovative approach that combines near-field spatial multiplexing with the interference mitigation capabilities of CRA, utilizing an extremely large aperture array at the access point. This integration improves reliability and reduces access latency, offering a robust framework for IIoT connectivity in next-generation 6G networks.

Figures

Figures reproduced from arXiv: 2508.15673 by the authors.

**Figure 2.** shows the PLR of the proposed CSRA scheme, both with and without SIC, as a function of the number of active NUs for various numbers of transmitted packet replicas, R. The results demonstrate the essential role of SIC in ensuring the reliability of the scheme. Without employing SIC, the system faces challenges in reaching even a packet loss probability of PL = 10−2 in the given scenario. The results also highlight th… view at source ↗

**Figure 3.** Packet loss rate PL of the proposed CSRA scheme as a function of the number of replicas R, with and without SIC. in [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Packet loss rate PL of the proposed CSRA scheme as a function of the number of active users K, with and without SIC. Comparison with single-element (SE) processing at the ELAA. C. Comparison with CSRA-SE Since the grant-free nature of this communication setup prevents …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 19 canonical work pages

[18]

Communication modes with large intelligent surfaces in the near field,

N. Decarli and D. Dardari, “Communication modes with large intelligent surfaces in the near field,”IEEE Access, vol. 9, pp. 165 648–165 666, Dec. 2021

work page 2021
[1]

Toward massive machine type cellular communications,

Z. Dawy, W. Saad, A. Ghosh, J. G. Andrews, and E. Yaacoub, “Toward massive machine type cellular communications,”IEEE Wireless Commun., vol. 24, no. 1, pp. 120–128, Nov. 2017

work page 2017
[2]

Massive machine-type communications in 5G: Physical and MAC-layer solutions,

C. Bockelmannet al., “Massive machine-type communications in 5G: Physical and MAC-layer solutions,”IEEE Commun. Mag., vol. 54, no. 9, pp. 59–65, Sep. 2016

work page 2016
[3]

Towards enabling critical mMTC: A review of URLLC within mMTC,

S. R. Pokhrel, J. Ding, J. Park, O.-S. Park, and J. Choi, “Towards enabling critical mMTC: A review of URLLC within mMTC,”IEEE Access, vol. 8, pp. 131 796–131 813, Jul. 2020

work page 2020
[4]

Access point cooperation strategies for coded random access in cell-free massive MIMO,

E. Testi, V . Tralli, and E. Paolini, “Access point cooperation strategies for coded random access in cell-free massive MIMO,”IEEE Internet Things J., vol. 11, no. 17, pp. 27 916–27 931, Jul. 2024

work page 2024
[5]

Coded random access: Applying codes on graphs to design random access protocols,

E. Paolini, ˇC. Stefanovi ´c, G. Liva, and P. Popovski, “Coded random access: Applying codes on graphs to design random access protocols,” IEEE Commun. Mag., vol. 53, no. 6, pp. 144–150, Jun. 2015

work page 2015
[6]

Holographic communication using intelligent surfaces,

D. Dardari and N. Decarli, “Holographic communication using intelligent surfaces,”IEEE Commun. Mag., vol. 59, no. 6, pp. 35–41, June 2021

work page 2021
[7]

A tutorial on near-field XL-MIMO communications towards 6G,

H. Luet al., “A tutorial on near-field XL-MIMO communications towards 6G,”IEEE Commun. Surv. Tutor ., pp. 1–1, Jan. 2024

work page 2024

Show all 20 references

[8]

A tutorial on extremely large-scale MIMO for 6G: Fundamentals, signal processing, and applications,

Z. Wanget al., “A tutorial on extremely large-scale MIMO for 6G: Fundamentals, signal processing, and applications,”IEEE Commun. Surv. Tutor ., vol. 26, no. 3, pp. 1560–1605, 2024

2024
[9]

Terahertz networks for future industrial internet of things,

S. Cavalleroet al., “Terahertz networks for future industrial internet of things,”ITU J. Future and Evolving Technologies, vol. 4, pp. 196–208, 03 2023

2023
[10]

Beam division multiple access transmission for massive MIMO communications,

C. Sunet al., “Beam division multiple access transmission for massive MIMO communications,”IEEE Trans. Commun., vol. 63, no. 6, pp. 2170–2184, Apr. 2015

2015
[11]

Multiple access for near-field communications: SDMA or LDMA?

Z. Wu and L. Dai, “Multiple access for near-field communications: SDMA or LDMA?”IEEE J. Sel. Areas Commun., vol. 41, no. 6, pp. 1918–1935, Jun. 2023

1918
[12]

Near-field rainbow: Wideband beam training for XL-MIMO,

M. Cui, L. Dai, Z. Wang, S. Zhou, and N. Ge, “Near-field rainbow: Wideband beam training for XL-MIMO,”IEEE Trans. Wireless Commun., vol. 22, no. 6, pp. 3899–3912, Jun. 2022

2022
[13]

Grant-free access for extra-large MIMO systems subject to spatial non-stationarity,

H. Iimoriet al., “Grant-free access for extra-large MIMO systems subject to spatial non-stationarity,” inProc. 2022 IEEE Int. Conf. Commun., Seoul, South Korea, May 2022, pp. 1758–1762

2022
[14]

QoS-aware user scheduling in crowded XL-MIMO systems under non- stationary multi-state LoS/NLoS channels,

J. H. I. de Souza, J. C. Marinello Filho, A. Amiri, and T. Abr ˜ao, “QoS-aware user scheduling in crowded XL-MIMO systems under non- stationary multi-state LoS/NLoS channels,”IEEE Trans. V ehicular Tech., vol. 72, no. 6, pp. 7639–7652, Jun. 2023

2023
[15]

Beamspace modulation for near field capacity improvement in XL-MIMO communications,

S. Guo and K. Qu, “Beamspace modulation for near field capacity improvement in XL-MIMO communications,”IEEE Wireless Commun. Lett., vol. 12, no. 8, pp. 1434–1438, May 2023

2023
[16]

Cell-free massive MIMO with radio stripes and sequential uplink processing,

Z. H. Shaik, E. Bj ¨ornson, and E. G. Larsson, “Cell-free massive MIMO with radio stripes and sequential uplink processing,” inProc. 2020 IEEE Int. Conf. Commun. Workshops, Online, Jun. 2020, pp. 1–6

2020
[17]

Balanis,Antenna Theory: Analysis and Design

C. Balanis,Antenna Theory: Analysis and Design. Wiley, 2015

2015
[19]

DFT-based beamforming weight- vector codebook design for spatially correlated channels in the unitary precoding aided multiuser downlink,

D. Yang, L.-L. Yang, and L. Hanzo, “DFT-based beamforming weight- vector codebook design for spatially correlated channels in the unitary precoding aided multiuser downlink,” inProc. 2010 IEEE Int. Conf. Commun., Cape Town, South Africa, May 2010, pp. 1–5

2010
[20]

Irregular repetition slotted ALOHA in an information-theoretic setting,

E. Paolini, L. Valentini, V . Tralli, and M. Chiani, “Irregular repetition slotted ALOHA in an information-theoretic setting,” inProc. 2022 IEEE Int. Symp. Inf. Theory, Espoo, Finland, Jun. 2022

2022

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.

[1] [18]

Communication modes with large intelligent surfaces in the near field,

N. Decarli and D. Dardari, “Communication modes with large intelligent surfaces in the near field,”IEEE Access, vol. 9, pp. 165 648–165 666, Dec. 2021

work page 2021

[2] [1]

Toward massive machine type cellular communications,

Z. Dawy, W. Saad, A. Ghosh, J. G. Andrews, and E. Yaacoub, “Toward massive machine type cellular communications,”IEEE Wireless Commun., vol. 24, no. 1, pp. 120–128, Nov. 2017

work page 2017

[3] [2]

Massive machine-type communications in 5G: Physical and MAC-layer solutions,

C. Bockelmannet al., “Massive machine-type communications in 5G: Physical and MAC-layer solutions,”IEEE Commun. Mag., vol. 54, no. 9, pp. 59–65, Sep. 2016

work page 2016

[4] [3]

Towards enabling critical mMTC: A review of URLLC within mMTC,

S. R. Pokhrel, J. Ding, J. Park, O.-S. Park, and J. Choi, “Towards enabling critical mMTC: A review of URLLC within mMTC,”IEEE Access, vol. 8, pp. 131 796–131 813, Jul. 2020

work page 2020

[5] [4]

Access point cooperation strategies for coded random access in cell-free massive MIMO,

E. Testi, V . Tralli, and E. Paolini, “Access point cooperation strategies for coded random access in cell-free massive MIMO,”IEEE Internet Things J., vol. 11, no. 17, pp. 27 916–27 931, Jul. 2024

work page 2024

[6] [5]

Coded random access: Applying codes on graphs to design random access protocols,

E. Paolini, ˇC. Stefanovi ´c, G. Liva, and P. Popovski, “Coded random access: Applying codes on graphs to design random access protocols,” IEEE Commun. Mag., vol. 53, no. 6, pp. 144–150, Jun. 2015

work page 2015

[7] [6]

Holographic communication using intelligent surfaces,

D. Dardari and N. Decarli, “Holographic communication using intelligent surfaces,”IEEE Commun. Mag., vol. 59, no. 6, pp. 35–41, June 2021

work page 2021

[8] [7]

A tutorial on near-field XL-MIMO communications towards 6G,

H. Luet al., “A tutorial on near-field XL-MIMO communications towards 6G,”IEEE Commun. Surv. Tutor ., pp. 1–1, Jan. 2024

work page 2024