Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

Ray Antenna Array: A Novel Cost-Effective Multi-Antenna Architecture for Enhanced Wireless Communication

T0 review · 4 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper claims that a ray antenna array of fixed tilted uniform linear subarrays, connected by switches, can outperform hybrid beamforming at 0.69% of the hardware cost with no phase shifters.

desk verdict RAA is a neat fixed-beam array concept with a credible cost story, but the headline 6 dB SNR gain is a product of asymmetric element-pattern comparisons; the architecture still merits peer review. read the letter →

arxiv 2505.18163 v1 pith:A4IZS6TX submitted 2025-05-13 eess.SP cs.ARcs.ITmath.IT

Zhenjun Dong , Zhiwen Zhou , Yong Zeng This is my paper · ORCID
classification eess.SPcs.ARcs.ITmath.IT
keywords rayantennaarrayuniformlinearhybridbeamformingphase-shifter-freeselectionnetworkcost-efficientMIMOmillimeter-wavecommunication
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 tries to show that beamforming can be done by geometry and switches instead of phase shifters. The ray antenna array (RAA) places many small uniform linear arrays, each tilted at a different fixed angle so that each one already points its beam where it is needed; a switch network selects the subarrays to connect to a few radio-frequency chains. The paper claims this reaches the same signal-combining gain as a DFT-codebook hybrid beamformer when a path arrives perpendicular to a selected ray, and that its uniform angular sampling gives more consistent coverage than HBF's $\sin\theta$ sampling. When the elements are made directional, the simulated single-user SNR rises by about 6 dB and multi-user sum rates improve, while the hardware cost is about 0.69% of the hybrid beamformer's. The payoff, if the claims hold, is a cheaper multi-antenna architecture that does not sacrifice performance.

What carries the argument

The load-bearing mechanism is the ray-like geometry: $N$ copies of an $M$-element uniform linear array (sULA) are tilted so that ray $n$ points at angle $\eta_n = n\arcsin(2/M)$ relative to the reference direction. This choice aligns each ray's main-lobe peak with the adjacent ray's first null, so the set of fixed beams forms a near-orthogonal tiling of the covered angular range. Each sULA is hardwired to an RF combiner, producing the Dirichlet-kernel response $H_M(\sin(\theta-\eta_n))$ --- the standard array gain pattern of a directly summed uniform line array --- and a ray selection network of switches connects the $N_{\mathrm{RF}}$ best rays to baseband. Selection thereby replaces phase-shifter steering: the physical tilt does the work that analog weights normally do.

What would settle it

Re-run the Section V simulations with the hybrid-beamforming baseline allowed the same directional element pattern as the RAA (or a lens-array/subarray baseline that can also use directional elements); if the SNR and sum-rate gaps nearly vanish, the performance advantage claimed for the architecture is not structural. A complementary check is to measure a physical RAA's beam pattern at $M=16$, $N=25$ and verify that the null-peak alignment at $\eta_n=n\arcsin(1/8)$ suppresses inter-ray interference as modeled.

Watch

Extended reading notes

Core claim

The central claim is that a fixed set of simple uniform linear arrays ('rays'), each with all elements hardwired to an RF combiner and no phase shifters, provides flexible beamforming by selection alone. Orient the $n$-th ray at $\eta_n = n\arcsin(2/M)$ with $M$ elements per ray. Because the nulls of each ray's main lobe fall at the peaks of its neighbors, adjacent rays interfere minimally and the assembly tiles the angular range $[-\eta_{\max},\eta_{\max}]$. Any direction in that range is served by picking the ray whose beam is closest; when the propagation path is exactly perpendicular to a selected ray, the array gives the full $M$-fold gain, the same as an ideal DFT-codebook beamformer. The paper further claims that because each ray covers only a slice of the angular range, its elements can be made directional; in the simulations this adds about 6 dB to the SNR and raises multi-user sum rates relative to a hybrid beamformer with broad elements, while the required hardware --- switches and antenna elements instead of phase shifters --- costs about 0.69% of the hybrid beamformer's hardware.

Load-bearing premise

The load-bearing premise is the antenna-pattern comparison: the RAA is assigned directional elements with a 3 dB beamwidth of $0.3\pi$ and a peak gain of about $5.13$ dB, while the hybrid-beamforming baseline is restricted to a broad element with $\theta_{3\mathrm{dB}}=\pi$ and $0$ dB peak gain, so much of the reported 6 dB SNR gain is built into that asymmetry.

Editorial extensions

If this is right

  • If the RAA claim holds, a fixed set of tilted subarrays plus a switch network reaches the same $M$-fold array gain as DFT-codebook hybrid beamforming whenever a propagation path is perpendicular to a selected ray, with no phase shifters.
  • Because each ray covers only part of the angular range, its elements can be more directional; the paper's simulations show this configuration adds roughly 6 dB to the SNR and raises multi-user sum rates against an HBF with broad elements.
  • The orientation rule $\eta_n = n\arcsin(2/M)$ gives uniform sampling of angle rather than of $\sin\theta$, so spatial resolution stays more consistent across the coverage range.
  • The greedy ray-selection algorithm closely matches exhaustive search in simulation, so the architecture remains practical when many rays and multiple RF chains are available.
  • In the quoted hardware-price example, the RAA's switches-plus-antennas bill is about 0.69% of the fully-connected HBF's phase-shifter-dominated bill.

Reading between the lines

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

  • Editorial inference: the same angle-uniform tiling could carry over to direction-of-arrival estimation and integrated sensing, where uniform angular resolution simplifies estimation; the paper only gestures at sensing without developing that use.
  • Editorial inference: a natural next comparison is to give the hybrid-beamforming baseline the same directional element pattern (or a lens-array equivalent); the current simulations do not include it, and running it would separate the geometric benefit of ray selection from the antenna-pattern gain.
  • Editorial inference: the selection-based steering idea could extend to wideband or very-large-array near-field settings, where ray orientations could be optimized per user location rather than fixed once for the whole angular range.
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.

Referee Report

4 major / 6 minor

Summary. The paper proposes a multi-antenna architecture called the ray antenna array (RAA), composed of N uniform linear subarrays (sULAs) of M elements each, arranged in a ray-like structure. Each sULA is connected to an RF combiner without phase shifters, and a ray selection network (RSN) connects a subset of NRF subarray outputs to baseband chains. The orientations are designed as η_n = n arcsin(2/M) so that the peak of one ray's main lobe aligns with the first null of the adjacent ray. The authors derive the input-output model, present joint ray selection and beamforming algorithms for single- and multi-user systems, and provide simulation results claiming that RAA achieves about 6 dB higher SNR than hybrid beamforming (HBF) and higher sum rates, while reducing hardware cost to roughly 6.9 per mille of HBF. The core architectural claim is that a fixed-beam, no-phase-shifter array can serve as a low-cost alternative to hybrid beamforming.

Significance. If the claims are substantiated, the RAA is a genuinely novel phase-shifter-free architecture with a clean theoretical foundation: the input-output model follows from the Dirichlet kernel, the orientation rule (6) is a transparent closed-form design, and the greedy ray-selection algorithm is validated against exhaustive search in simulation. The cost model is also explicitly stated and transparent. However, the headline performance advantage over HBF is not yet established. The simulation comparison in Sec. V gives the RAA directional elements (θ3dB=0.3π, G0=5.1335 dB) while restricting HBF to isotropic elements (θ3dB=π, G0=0 dB), so the approximately 6 dB gain is a consequence of this asymmetry rather than of the ray architecture itself. The comparison also involves an unequal number of antennas (400 vs. 16 in the default setting). A fair baseline with directional elements, multiple subarrays, or the prior lens-array architecture [9] is needed. The architecture and derivations are sound enough to merit revision; the performance claims need recalibration.

major comments (4)
  1. [Sec. V, Eq. (15), Fig. 4] The claimed SNR advantage of the RAA is produced by an asymmetric element-pattern setup: the RAA elements are assigned θ3dB=0.3π and G0=5.1335 dB, while the HBF baseline is restricted to θ3dB=π and G0=0 dB. Since Fig. 4 shows essentially no SNR difference with isotropic elements, the approximately 6 dB gain in Fig. 4 is attributable to the directional-element allocation, not to the ray architecture itself. A fair baseline, such as an HBF with multiple directionally oriented subarrays or the lens antenna array of [9], is required before the stated performance advantage over HBF can be accepted.
  2. [Sec. III and Sec. V] The comparison is also unequal in antenna resources: the RAA uses NM elements (400 for the M=16, N=25 setting), whereas the HBF baseline uses M=16 elements. Any architecture with more elements and narrower per-array coverage could exhibit a similar gain, so the comparison should specify whether it is equal element count, equal aperture, or equal hardware cost, and ideally include baselines at the same resource budget.
  3. [Sec. III, hardware cost example] The reported cost ratio is inconsistent with the stated formula and unit prices. For N=25, M=16, NRF=1, psw=0.12$ and pant=0.01$, Eq. cost_RAA = NRF*N*psw + N*M*pant gives 3.16$, not the stated 7$, so the relative cost is about 3.1 per mille of the HBF cost, not 6.9 per mille. Please correct the arithmetic and the resulting ratio.
  4. [Sec. V, Table I, Fig. 5] The multi-user simulation places user mean angles uniformly in θ (θ̄_k = −0.5π + 0.15π k), which matches the RAA's uniform-θ sampling and is unfavorable to the DFT codebook's uniform-sinθ sampling. The sum-rate advantage in Fig. 5 is therefore partly a consequence of the user distribution; results for users uniform in sinθ or other distributions should be reported before claiming a general performance enhancement.
minor comments (6)
  1. [Fig. 3 and Fig. 4] The beam pattern for the RAA is not defined; please state whether the plotted quantity is the maximum over all sULAs, the pattern of the best selected ray, or something else.
  2. [Sec. IV, Algorithm 1] The complexity is stated as O(N NRF), but each candidate evaluation involves matrix inversions and sums of logarithms; a more explicit complexity analysis including the dimensions NRF and K is needed.
  3. [Sec. II] The condition on D ensures separation of the first elements of adjacent sULAs, but the manuscript does not discuss whether other elements of neighboring sULAs may overlap or couple; a brief layout feasibility argument would be helpful.
  4. [Sec. II] The symbol N is used both for the set of sULA indices and for the number of sULAs; using a calligraphic symbol for the index set would remove ambiguity.
  5. [Eq. (2) and (3)] The notation b is used for both the vector in (2) and the scalar element pattern in (3); please use distinct symbols or explicitly define the vector entries.
  6. [Sec. III, cost comparison] The cost comparison does not include the insertion loss and power consumption of the RSN switches or the RF combiners; a note on these overheads would strengthen the cost claim.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: RAA beamforming and orientation derivations are self-contained; the 6 dB SNR gain is an explicit, unfitted consequence of the directional-element simulation setup.

full rationale

The derivation chain is self-contained and not circular. The RAA beamforming property follows directly from the array response model in (1)-(5): a directly-connected M-element sULA has a Dirichlet-kernel response peaked at the physical orientation θ=η_n, so no phase shifters are needed. The ray orientations (6) are chosen from an explicit null-alignment criterion, not from the target performance. The M-fold array gain when θ=η_n is derived from the same model, and the comparison with DFT-codebook HBF in (7) is an analytic comparison of the two array responses. No parameter is fitted to a target result, and no uniqueness theorem or load-bearing conclusion is imported from the authors' prior work; the self-citations [1,2,4,5,9,12,13] are background only. The only notable caveat is that the roughly 6 dB SNR advantage in Fig. 4 is a direct consequence of the element-pattern parameters chosen in (15) (G0=5.1335 dB, θ3dB=0.3π for RAA versus G0=0 dB, θ3dB=π for HBF), and the paper explicitly attributes the gain to this directional-element assumption. That makes the comparison sensitive to baseline fairness, but it is not circular: the assumption is stated, not fitted, and the analytic SNR expression (12) is evaluated rather than reverse-engineered.

Assumptions & free parameters 7 free parameters · 5 assumptions · 1 invented entities

The central model is self-contained and has no fitted constants, but the headline performance comparison rests on several hand-chosen parameters, including element beamwidths and gains, and on ideal-hardware assumptions that are not independently verified.

free parameters (7)
  • M = 16
    Number of elements per sULA; with eta_max = 0.5 pi it sets N=25 and the per-ray beamwidth. A system design choice, not fitted to data.
  • NRF = 5
    Number of RF chains and selected rays in simulations; the sum-rate results scale with this choice.
  • eta_max = 0.5 pi
    Maximum ray orientation; defines the coverage angle and, with M, the number of rays N.
  • theta_3dB_RAA = 0.3 pi
    3 dB beamwidth of RAA antenna elements in Eq. (15), Sec. V. Narrower than HBF's element, this setting produces the directional gain advantage in Figs. 3 to 5.
  • G0_RAA = 5.1335 dB
    Peak gain of RAA elements, chosen so that total radiated power equals HBF's. It directly translates the narrow beamwidth into higher on-boresight gain.
  • theta_3dB_HBF = pi
    Broad element pattern assigned to the HBF baseline. This choice makes the HBF lower-gain off-boresight and is not tested for sensitivity.
  • G0_HBF = 0 dB
    Peak gain of the HBF baseline element; with theta_3dB = pi it approximates an isotropic element. The fairness of this baseline is a key assumption.
assumptions (5)
  • domain assumption Far-field narrowband plane-wave channel model with array response in Eqs. (1) to (3).
    All derivations of beam patterns and SNR assume plane-wave propagation and no mutual coupling between elements or rays.
  • domain assumption Radiation pattern of each element is G(theta - eta_n) with boresight along the ray orientation.
    The model treats every element in an sULA as having the same directional pattern aligned with the ray; physical element mounting and mutual coupling may break this.
  • ad hoc to paper Interference between adjacent sULAs is minimized by aligning the peak of one main lobe with the first null of the adjacent main lobe, Eq. (6).
    This criterion is stated as the design goal, not derived from an optimality condition; with directional element patterns the actual nulls shift, so the alignment is approximate.
  • ad hoc to paper Equal total radiated power between RAA and HBF is ensured by choosing G0_RAA = 5.1335 dB with theta_3dB_RAA = 0.3 pi.
    This fairness constraint is introduced in Sec. V and is what converts the narrower RAA beamwidth into a 6 dB peak-gain advantage.
  • domain assumption HBF baseline uses a DFT codebook with N' = 16 codewords and a single M-element ULA.
    The comparison is limited to fully-connected HBF with one subarray; other HBF variants with directional elements or subarrays are not considered.
invented entities (1)
  • Ray Antenna Array (RAA)
    purpose: A multi-antenna architecture using M-element sULAs in ray orientations with direct RF combiners and a switch-based ray selection network, to form beams without phase shifters.
    The paper provides simulations but no prototype, measured radiation patterns, or hardware experiments; the only evidence is the analytical model and simulated beam patterns.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Ray Antenna Array: A Novel Cost-Effective Multi-Antenna Architecture for Enhanced Wireless Communication." pith.science (2026). https://pith.science/paper/A4IZS6TX

@misc{pith2026250518163,
  author       = {Pith},
  title        = {Pith review of: Ray Antenna Array: A Novel Cost-Effective Multi-Antenna Architecture for Enhanced Wireless Communication},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A4IZS6TX}},
  note         = {Machine review of arXiv:2505.18163}
}
read the original abstract

This paper proposes a novel multi-antenna architecture, termed ray antenna array (RAA), which aims to enhance wireless communication performance in a cost-effective manner. RAA is composed of massive cheap antenna elements and a few radio frequency (RF) chains. The massive antenna elements are arranged in a novel ray-like structure, with each ray corresponding to a simple uniform linear array (sULA) with a carefully designed orientation. The antenna elements of each sULA are directly connected to an RF combiner, so that the sULA in each ray is able to form a beam towards a direction matching the ray orientation without relying on any analog or digital beamforming. By further designing a ray selection network (RSN), appropriate sULAs are selected to connect to the RF chains for further baseband processing. Compared to conventional multi-antenna architectures like hybrid analog/digital beamforming (HBF), the proposed RAA has two major advantages. First, it can significantly reduce hardware costs since no phase shifters, which are usually expensive especially in high-frequency systems, are required. Besides, RAA can greatly improve system performance by configuring antenna elements with higher directionality, as each sULA only needs to be responsible for a portion of the total coverage angle. To demonstrate such advantages, in this paper, we first present the input-output model for RAA-based wireless communications, based on which the ray orientations of the RAA are designed. Furthermore, efficient algorithms for joint ray selection and beamforming are proposed for single-user and multi-user RAA-based wireless communications. Simulation results demonstrate the superior performance of RAA compared to HBF while significantly reducing hardware cost.

Figures

Figures reproduced from arXiv: 2505.18163 by the authors.

Figure 1
Figure 1. The proposed RAA consists of MN cost-effective antenna elements, which are arranged into N rays. Each ray corresponds to an M-element sULA, where all antenna elements are directly connected. As illustrated in [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 5
Figure 5. illustrates the achievable communication sum rates Rsum (bps/Hz) in (13) obtained via exhaustive search and the proposed greedy scheme for the RAA and HBF, considering both the isotropic and directional antenna elements. It can be observed from [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 3
Figure 3. Beam patterns of the proposed RAA and HBF, consid [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Maximum SNR (dB) in (12) for the RAA and HBF. [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Full-Angle Ray Antenna Array and Omnicell Wireless Communication System

    eess.SP 2025-09 reject novelty 4.0 of 10

    Full-angle RAA, many cheap direct-connected linear subarrays pointing in all directions, enables an 'omnicell' base station with uniform angular resolution and, in the paper's simulations, higher sum rate and lower ha...

Reference graph

Works this paper leans on

16 extracted references · 12 canonical work pages · cited by 1 Pith paper

  1. [9]

    Millimeter wave MIMO with lens antenna array: A new path division multiplexing paradigm,

    Y . Zeng and R. Zhang, “Millimeter wave MIMO with lens antenna array: A new path division multiplexing paradigm,” IEEE Trans. Commun. , vol. 64, no. 4, pp. 1557–1571, 2016

    work page 2016

  2. [1]

    An overview on integrated localization and communication towards 6G,

    Z. Xiao and Y . Zeng, “An overview on integrated localization and communication towards 6G,” Sci. China Inf. Sci. , vol. 65, no. 3, pp. 1–46, 2022

    work page 2022

  3. [2]

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

    H. Lu, Y . Zeng, C. You et al. , “A tutorial on near-field XL-MIMO communications towards 6G,” IEEE Commun. Surveys Tuts. , vol. 26, no. 4, pp. 2213–2257, 2024

    work page 2024

  4. [3]

    Cell-free massive MIMO versus small cells,

    H. Q. Ngo, A. Ashikhmin, H. Yang et al. , “Cell-free massive MIMO versus small cells,” IEEE Trans. Wireless Commun. , vol. 16, no. 3, pp. 1834–1850, 2017

    work page 2017

  5. [4]

    Sparse MIMO for ISAC: New opportunities and challenges,

    X. Li, H. Min, Y . Zeng et al. , “Sparse MIMO for ISAC: New opportunities and challenges,” IEEE Wireless Commun. Mag. , 2024, doi:10.1109/MWC.001.2400201

    doi:10.1109/mwc.001.2400201 2024

  6. [5]

    A tutorial on environment-aware communications via channel knowledge map for 6G,

    Y . Zeng, J. Chen, J. Xu et al. , “A tutorial on environment-aware communications via channel knowledge map for 6G,” IEEE Commun. Surveys Tuts, no. 3, pp. 1478–1519, 2024

    work page 2024

  7. [6]

    Beam codebook based beamforming protocol for multi-Gbps millimeter-wave WPAN systems,

    J. Wang, Z. Lan, C.-w. Pyo et al., “Beam codebook based beamforming protocol for multi-Gbps millimeter-wave WPAN systems,” IEEE J. Sel. Areas Commun., vol. 27, no. 8, pp. 1390–1399, 2009

    work page 2009

  8. [7]

    Spatially sparse precoding in millimeter wave MIMO systems,

    O. El Ayach, S. Rajagopal, S. Abu-Surra et al. , “Spatially sparse precoding in millimeter wave MIMO systems,” IEEE Trans. Wireless Commun., vol. 13, no. 3, pp. 1499–1513, 2014

    work page 2014

Show all 16 references
  1. [8]

    Hybrid MIMO architectures for millimeter wave communications: Phase shifters or switches?

    R. M ´endez-Rial, C. Rusu, N. Gonz ´alez-Prelcic et al. , “Hybrid MIMO architectures for millimeter wave communications: Phase shifters or switches?” IEEE Access , vol. 4, pp. 247–267, 2016

    2016

  2. [10]

    Fluid antenna systems,

    K.-K. Wong, A. Shojaeifard, K.-F. Tong et al., “Fluid antenna systems,” IEEE Trans. Wireless Commun. , vol. 20, no. 3, pp. 1950–1962, 2020

    1950

  3. [11]

    Movable antennas for wireless commu- nication: Opportunities and challenges,

    L. Zhu, W. Ma, and R. Zhang, “Movable antennas for wireless commu- nication: Opportunities and challenges,” IEEE Commun. Mag. , vol. 62, no. 6, pp. 114–120, 2024

    2024

  4. [12]

    Movable antenna for wireless com- munications: Prototyping and experimental results,

    Z. Dong, Z. Zhou, Z. Xiao et al. , “Movable antenna for wireless com- munications: Prototyping and experimental results,” submitted to IEEE Trans. Wireless Commun., 2024. Avaliable online:arXiv:2408.08588

    2024 arXiv

  5. [13]

    Group movable antenna with flexible sparsity: Joint array position and sparsity optimization,

    H. Lu, Y . Zeng, S. Jin et al. , “Group movable antenna with flexible sparsity: Joint array position and sparsity optimization,” IEEE Wireless Commun. Lett. , vol. 13, no. 12, pp. 3573–3577, 2024

    2024

  6. [14]

    S. U. Pillai, Array signal processing . Springer Science & Business Media, 2012

    2012

  7. [15]

    Qorvo official website,

    Qorvo, “Qorvo official website,” https://www.qorvo.com/, 2025, ac- cessed: 2025-01-07

    2025

  8. [16]

    5G; Study on Channel Model for Frequencies from 0.5 to 100 GHz , document 38.901, Version 16.1.0, 3GPP, Technical Specification (TS), Nov. 2020

    2020

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.