Pith. sign in

REVIEW 2 major objections 4 minor 1 cited by

Capacity of MIMO Systems Aided by Microwave Linear Analog Computers (MiLACs)

T0 review · 2 major / 4 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 A pair of lossless, reciprocal microwave analog computers can match the capacity of fully digital MIMO beamforming with the same number of streams; the optimal network settings are given in closed form.

desk verdict Solid, useful result on lossless reciprocal MiLACs achieving digital MIMO capacity, but the upper-bound proof in Sec. IV-C uses a relaxation it does not actually solve; the fix is straightforward and the closed-form admittances need an invertibility caveat. read the letter →

arxiv 2506.05983 v1 pith:5WSE3U73 submitted 2025-06-06 cs.IT eess.SPmath.IT

Matteo Nerini , Bruno Clerckx This is my paper
classification cs.ITeess.SPmath.IT MSC 94A0594A2494A40
keywords microwavelinearanalogcomputer(MiLAC)giganticMIMOcapacitylosslessreciprocalnetworkbeamformingratemaximizationwater-fillingpowerallocationscatteringparameters
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 whether a MIMO link whose precoding and combining are done entirely by passive microwave circuits, not by digital signal processing, can still transmit at the information-theoretic rate. The answer it tries to establish is yes: with a MiLAC at both ends, restricted to lossless reciprocal components (pure reactances, no resistors, no non-reciprocal elements), the optimal analog networks are given in closed form as functions of the channel's singular vectors, and the resulting rate equals the capacity of a digital system sending the same number of streams. The capacity expression is $C = \sum_{s=1}^{N_S} \log_2\left(1 + \frac{P_T p_s \lambda_s}{4\sigma^2}\right)$ with water-filling power, so the only penalty relative to textbook MIMO capacity is the factor 4 in the SNR from voltage-divider coupling at the antenna ports. If right, this makes gigantic MIMO cheaper: only $N_S$ RF chains, low-resolution ADCs/DACs, and no per-symbol matrix-vector product, with no loss of capacity.

What carries the argument

The central object is the scattering matrix $\Theta$ of the MiLAC, related to its admittance matrix by $\Theta = (Y_0 I + Y)^{-1}(Y_0 I - Y)$; for a lossless reciprocal network $\Theta$ is symmetric unitary. Because the precoding matrix is just half a block of $\Theta$, the physical-network constraint becomes the requirement that a symmetric unitary matrix contain $\bar{V}$ (or $\bar{U}^H$) as a block. The paper's mechanism is a symmetric-unitary matrix completion: it writes $\Theta_F^\star = \begin{bmatrix} 0 & \bar{V}^T \\ \bar{V} & -\tilde{V}\tilde{V}^T \end{bmatrix}$ and checks unitarity and symmetry, turning the SVD/water-filling solution of the relaxed problem into a feasible analog network. The closed-form admittance matrices $B_F^\star$ and $B_G^\star$ follow by inverting the $Y \leftrightarrow \Theta$ relation with a $2 \times 2$ block inverse, expressed through $\Im\{V\}^{-1}$ and $\Im\{U\}^{-1}$. This is what carries the argument from the relaxed digital-domain optimum to a physically constrained analog realization.

What would settle it

Take a channel whose right singular vector matrix $V$ is real orthogonal, so $\Im\{V\} = 0$; the closed-form susceptance $B_F^\star$ in (81) involves $\Im\{V\}^{-1}$ and is undefined, and even in a limiting sense the required susceptance values diverge. If no other lossless reciprocal tuning achieves $C$ for such a channel, capacity equality fails. Equivalently, a circuit-level simulation of the network realizing (67)-(68) with ideal lossless reciprocal components should reproduce $C$ at every SNR; a measured gap larger than numerical error would falsify the claim.

Watch

Extended reading notes

Core claim

The central discovery is that for a point-to-point MIMO channel with $N_S$ streams, $N_S \le \min\{N_T, N_R\}$, a lossless reciprocal MiLAC at the transmitter with scattering matrix $\Theta_F^\star = \begin{bmatrix} 0 & \bar{V}^T \\ \bar{V} & -\tilde{V}\tilde{V}^T \end{bmatrix}$ and at the receiver with $\Theta_G^\star = \begin{bmatrix} -\tilde{U}^*\tilde{U}^H & \bar{U}^* \\ \bar{U}^H & 0 \end{bmatrix}$ implements the precoder $F^\star = \bar{V}/2$ and combiner $G^\star = \bar{U}^H/2$. These are symmetric unitary matrices, hence realizable by ideal lossless reciprocal susceptance networks. The resulting effective channel $G^\star H F^\star$ is diagonal with entries $\sigma_s/4$, and with water-filling over the $N_S$ strongest singular values the rate reaches $C = \sum_{s=1}^{N_S} \log_2\left(1 + \frac{P_T p_s \lambda_s}{4\sigma^2}\right)$. The paper proves that this closed-form solution is feasible, because it satisfies the symmetric-unitary constraints, and globally optimal even though the original problem is non-convex, by first solving a relaxed upper-bound problem and then showing the relaxation is tight at the optimum.

Load-bearing premise

The load-bearing premise is that each tunable component behaves as an ideal lossless reciprocal element with zero resistance, unlimited tuning range, and perfectly matched ports, so the network can realize exactly the symmetric unitary matrices (67) and (68); any parasitic loss or mismatch breaks the equality between analog and digital capacity.

Editorial extensions

If this is right

  • Lossless and reciprocal MiLACs attain the same capacity as digital beamforming with the same number of streams, so the analog architecture does not trade capacity for cost.
  • The required number of RF chains drops to $N_S$ at both transmitter and receiver, the minimum needed to carry $N_S$ streams.
  • ADCs and DACs can use lower resolution because each RF chain carries one symbol stream rather than a linear combination of all streams.
  • Precoding and combining are performed in the analog domain, eliminating the per-symbol matrix-vector multiplication.
  • The optimal susceptance matrices are directly computable from the channel SVD, so the network can be reconfigured from estimated channel state information.

Reading between the lines

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

  • Because the proof assumes ideal lossless reciprocal components, a natural next test is to simulate the same networks with realistic varactor losses and limited tuning ranges; the predicted capacity would stop matching digital beamforming once parasitic resistance or component mismatch becomes significant.
  • The paper fixes $N_S$ and notes the expression equals information-theoretic capacity only when $\mathrm{rank}(H) = N_S$; allowing the number of streams to be optimized would require comparing MiLAC-aided systems against the full capacity with optimized stream count, which is not resolved here.
  • The closed-form completion technique for symmetric unitary matrices may transfer to multi-user or distributed settings where the same block structure appears, though the paper does not analyze those cases.
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

2 major / 4 minor

Summary. The paper studies a point-to-point MIMO system in which both the transmitter and the receiver use microwave linear analog computers (MiLACs) to perform beamforming entirely in the analog domain. The MiLACs are constrained to be lossless and reciprocal, and the authors formulate a rate maximization problem over the susceptance matrices of the two networks and the power allocation. After translating the constraints into scattering-parameter form, they relax the problem, invoke a standard water-filling argument, and claim a closed-form global solution whose achieved rate equals the digital beamforming capacity with the same number of streams, C = sum_s log2(1 + PT p_s λ_s/(4σ^2)). Sections IV and V construct symmetric unitary scattering matrices whose sub-blocks are the dominant singular vectors of the channel, and Section V-C converts them back to closed-form susceptance expressions. The paper concludes that lossless reciprocal MiLACs achieve the same capacity as digital beamforming while using only NS RF chains.

Significance. If the result is correct, it is a significant advance for gigantic MIMO: it shows that a fully analog, lossless, and reciprocal microwave network can implement capacity-optimal precoding and combining with only one RF chain per stream, eliminating the usual digital matrix-vector multiplication and high-resolution ADCs/DACs. The key construction—embedding the optimal SVD precoder/combiner into symmetric unitary scattering matrices—is elegant, and the derivation is self-contained, uses no fitted parameters, and provides closed-form expressions. The numerical comparison between MiLAC, digital, and the claimed capacity is a useful sanity check. However, as detailed below, the proof that the water-filling expression is an upper bound on the original problem is flawed as written, so the central claim of global optimality is not yet established by the manuscript.

major comments (2)
  1. [Section IV-C, Eqs. (47)-(52)] The relaxation used to obtain the water-filling solution is incorrect. The paper replaces the per-column unit-norm constraints on Fbar with the single Frobenius-norm constraint ||Fbar||_F^2 ≤ NS, but the standard water-filling result cited from [30, Chapter 5] applies to a total-power constraint on the covariance Q = Fbar P Fbar^H, i.e., Tr(Q) ≤ 1. Under (48), Tr(Q) can be as large as NS because an individual column of Fbar may have norm up to sqrt(NS). For example, with NS = 2 and two equal singular values λ, the claimed optimum Fbar = [v1, v2], P = diag(1/2,1/2) gives 2 log(1 + cλ/2), while the feasible point Fbar = [sqrt(2) v1, 0], P = diag(1,0) satisfies ||Fbar||_F^2 = 2 and yields log(1 + 2cλ), which is larger for cλ = 0.1. Thus the maximum of the relaxed problem (47)-(49) is not the C given in (52), and the paper has not shown that C is an upper bound on the original rate. The original constraints do imply ||fbar_s|| ≤ 1 for each column, which gives Tr(Q) ≤ 1, so the intended bound can be recovered with a different relaxation; but the proof as written uses the wrong one and this is load-bearing for the global-optimality claim in Section V-B.
  2. [Section V-C, Eqs. (81) and (93)] The closed-form susceptance expressions require the imaginary parts of the singular-vector matrices, ℑ{V} and ℑ{U}, to be invertible. This excludes real-valued channels, for which the right singular vectors can be chosen real and ℑ{V} = 0, making the formulas undefined. The paper states the invertibility condition inline, but the abstract and contributions claim a global closed-form solution without this caveat. The authors should either provide a limiting or alternative construction for the singular case (e.g., using the alternative completions mentioned in the footnote to (67) or a continuity argument) or explicitly state the domain of validity of the closed-form formulas.
minor comments (4)
  1. [General notation] The paper uses the term 'capacity' for the maximum rate with a fixed number of streams NS. Since the information-theoretic capacity requires optimizing over NS (as the authors note in Section VI), the abstract and introduction should qualify that this is the capacity with a given number of spatial streams.
  2. [Section V-C, Eq. (70)] The scattering-to-admittance conversion involves (ΘF + I)^{-1}. For some channels, e.g., NS = NT with a real channel, the optimal ΘF in (67) can have an eigenvalue equal to −1, making this inverse singular and the admittance representation ill-defined. The paper should discuss this physical realizability limitation or state how it is handled in practice.
  3. [Section VI and Appendix] The derivation of the digital capacity uses the factor 1/4 arising from the antenna–RF chain voltage dividers. This is a correct and important detail, but it appears only in the Appendix; a cross-reference in Section VI would help the reader.
  4. [Figures 7 and 8] The legends in the figures are a bit crowded and the label 'C' is not defined in the caption; the caption should state that C is the expression in (52), and the curves for different antenna counts could be distinguished more clearly.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the MiLAC capacity result is derived from standard MIMO theory with an explicit feasibility construction, and prior self-citations are not load-bearing for the capacity claim.

full rationale

The derivation chain is self-contained and does not reduce to its own inputs. The capacity claim is obtained by (i) modeling the MiLAC precoder and combiner through the standard scattering-parameter map F = (1/2)[Theta_F] and G = (1/2)[Theta_G] in (31) and (33), (ii) upper bounding the per-stream rate by the mutual information I(s,y) in (42), (iii) relaxing the unitary-block constraint to ||Fbar||_F^2 <= NS in (47)-(49), (iv) invoking the classical water-filling solution of the relaxed problem to obtain C in (52), and (v) proving in Section V that the water-filling point can be completed to symmetric unitary matrices Theta_F and Theta_G, with closed-form susceptance matrices B_F and B_G in (81) and (93), and that at this point the rate in (69) equals the bound in (52). The equality with digital beamforming is then a comparison against an independently derived expression (97) with the same water-filling allocation. No parameter is fitted to the data being predicted, and no uniqueness or feasibility claim is imported from the authors' prior work: the self-citations [24] and [25] only supply the architectural model of the MiLAC and its circuit-theoretic admittance description, which is restated as a premise rather than used as a substitute for the capacity proof. A possible technical concern about whether (48) is the exact relaxation solved by the cited water-filling theorem is a proof-correctness issue, not a circularity: it does not make the claimed result equivalent by construction to its assumptions.

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

The central claim rests on standard MIMO and microwave theory plus the physical realizability of MiLAC from prior work. No fitted parameters are introduced; the power allocation is optimized, not fitted.

assumptions (5)
  • domain assumption Any symmetric unitary scattering matrix can be realized by a lossless reciprocal multiport network built from tunable admittance components
    Assumed from prior work [24], [25]; not proven in this paper. Used in Section V-A to assert that the constructed ΘF and ΘG are physically implementable.
  • domain assumption Antennas and RF chains are perfectly matched to reference impedance Z0
    Used in Section VI and the Appendix to relate generator signals to antenna signals with a 1/2 factor, which is essential for the fair digital comparison.
  • standard math Standard MIMO capacity results: SVD water-filling is optimal for the relaxed problem
    Invoked in Section IV-C, quoting [30, Chapter 5].
  • standard math Lossless microwave network has unitary scattering matrix; reciprocal network has symmetric scattering matrix
    Standard microwave theory from [28, Chapter 4], used in Section IV-A.
  • domain assumption The channel H is perfectly known at both ends
    Required to compute SVD and set the MiLAC parameters; not stated explicitly but implicit in the design.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Capacity of MIMO Systems Aided by Microwave Linear Analog Computers (MiLACs)." pith.science (2026). https://pith.science/paper/5WSE3U73

@misc{pith2026250605983,
  author       = {Pith},
  title        = {Pith review of: Capacity of MIMO Systems Aided by Microwave Linear Analog Computers (MiLACs)},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5WSE3U73}},
  note         = {Machine review of arXiv:2506.05983}
}
read the original abstract

Future wireless systems, known as gigantic multiple-input multiple-output (MIMO), are expected to enhance performance by significantly increasing the number of antennas, e.g., a few thousands. To enable gigantic MIMO overcoming the scalability limitations of digital architectures, microwave linear analog computers (MiLACs) have recently emerged. A MiLAC is a multiport microwave network that processes input microwave signals entirely in the analog domain, thereby reducing hardware costs and computational complexity of gigantic MIMO architectures. In this paper, we investigate the fundamental limits on the rate achievable in MiLAC-aided MIMO systems. We model a MIMO system employing MiLAC-aided beamforming at the transmitter and receiver, and formulate the rate maximization problem to optimize the microwave networks of the MiLACs, which are assumed lossless and reciprocal for practical reasons. Under the lossless and reciprocal constraints, we derive a global optimal solution for the microwave networks of the MiLACs in closed form. In addition, we also characterize in closed-form the capacity of MIMO systems operating MiLAC-aided beamforming. Our theoretical analysis, confirmed by numerical simulations, reveals that MiLAC-aided beamforming achieves the same capacity as digital beamforming, while significantly reducing the number of radio frequency (RF) chains, analog-to-digital converters (ADCs)/digital-to-analog converters (DACs) resolution requirements, and computational complexity.

Figures

Figures reproduced from arXiv: 2506.05983 by the authors.

Figure 1
Figure 1. Representation of (a) digital, (b) hybrid, (c) tri-hybrid, and (d) MiLAC [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. MIMO system with MiLAC at both the transmitter and receiver. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Example of lossless and reciprocal MiLAC at the transmitter, with [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 5
Figure 5. Figure 5: Framework for globally solving the rate maximization problem in [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Digital MIMO system. that can be achieved with MiLAC-aided beamforming. In this section, we relate the capacity achievable with MiLAC-aided beamforming to that achievable with digital beamforming. Consider a point-to-point MIMO system between an NT - antenna transmitte…
Figure 8
Figure 8. Figure 8: Achievable rate versus the number of antennas [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: Single-antenna transmitter and receiver in (a) and (c), respectively, [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]

Discussion (0). Sign in 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. Microwave Linear Analog Computers (MiLACs) for Communications: Opportunities and Challenges

    cs.IT 2026-07 conditional novelty 2.0 of 10

    Analog microwave networks can compute beamforming matrices, including matrix inversion, in one propagation step, promising quadratic instead of cubic complexity for massive-MIMO precoding.

Reference graph

Works this paper leans on

34 extracted references · 20 canonical work pages · cited by 1 Pith paper

  1. [1]

    Massive MIMO for next generation wireless systems,

    E. G. Larsson, O. Edfors, F. Tufvesson, and T. L. Marzetta, “Massive MIMO for next generation wireless systems,” IEEE Commun. Mag. , vol. 52, no. 2, pp. 186–195, 2014

    2014

  2. [2]

    Massive MIMO: Ten myths and one critical question,

    E. Bj ¨ornson, E. G. Larsson, and T. L. Marzetta, “Massive MIMO: Ten myths and one critical question,” IEEE Commun. Mag. , vol. 54, no. 2, pp. 114–123, 2016

    work page 2016

  3. [3]

    Enabling 6G performance in the upper mid-band by transitioning from massive to gigantic MIMO,

    E. Bj ¨ornson, F. Kara, N. Kolomvakis, A. Kosasih, P. Ramezani, and M. B. Salman, “Enabling 6G performance in the upper mid-band by transitioning from massive to gigantic MIMO,” arXiv preprint arXiv:2407.05630, 2024

    arXiv 2024

  4. [4]

    (2022) Vision, market drivers, and research directions on the path to 6G

    Qualcomm. (2022) Vision, market drivers, and research directions on the path to 6G. [Online]. Available: https://www.qualcomm.com/research/6g

    2022

  5. [5]

    MIMO for millimeter-wave wireless communications: Beamforming, spatial multiplexing, or both?

    S. Sun, T. S. Rappaport, R. W. Heath, A. Nix, and S. Rangan, “MIMO for millimeter-wave wireless communications: Beamforming, spatial multiplexing, or both?” IEEE Commun. Mag. , vol. 52, no. 12, pp. 110– 121, 2014

    work page 2014

  6. [6]

    Spatially sparse precoding in millimeter wave MIMO systems,

    O. E. Ayach, S. Rajagopal, S. Abu-Surra, Z. Pi, and R. W. Heath, “Spatially sparse precoding in millimeter wave MIMO systems,” IEEE Trans. Wireless Commun. , vol. 13, no. 3, pp. 1499–1513, 2014

    work page 2014

  7. [7]

    Hybrid digital and analog beamforming design for large-scale antenna arrays,

    F. Sohrabi and W. Yu, “Hybrid digital and analog beamforming design for large-scale antenna arrays,” IEEE J. Sel. Top. Signal Process., vol. 10, no. 3, pp. 501–513, 2016

    work page 2016

  8. [8]

    Embracing reconfigurable antennas in the tri-hybrid MIMO architecture for 6G,

    M. R. Castellanos, S. Yang, C.-B. Chae, and R. W. Heath Jr, “Embracing reconfigurable antennas in the tri-hybrid MIMO architecture for 6G,” arXiv preprint arXiv:2501.16610 , 2025. 13

    arXiv 2025

Show all 34 references
  1. [9]

    Intelligent reflecting surface-aided wireless communications: A tutorial,

    Q. Wu, S. Zhang, B. Zheng, C. You, and R. Zhang, “Intelligent reflecting surface-aided wireless communications: A tutorial,” IEEE Trans. Commun., vol. 69, no. 5, pp. 3313–3351, 2021

    2021

  2. [10]

    Electrically steerable passive array radiator (ESPAR) antennas,

    H. Kawakami and T. Ohira, “Electrically steerable passive array radiator (ESPAR) antennas,” IEEE Antennas Propag. Mag. , vol. 47, no. 2, pp. 43–50, 2005

    2005

  3. [11]

    Fluid antenna systems,

    K.-K. Wong, A. Shojaeifard, K.-F. Tong, and Y . Zhang, “Fluid antenna systems,” IEEE Trans. Wireless Commun., vol. 20, no. 3, pp. 1950–1962, 2021

    1950

  4. [12]

    Modeling and performance analysis for movable antenna enabled wireless communications,

    L. Zhu, W. Ma, and R. Zhang, “Modeling and performance analysis for movable antenna enabled wireless communications,” IEEE Trans. Wireless Commun., vol. 23, no. 6, pp. 6234–6250, 2024

    2024

  5. [13]

    Intelligent surface-aided transmitter architectures for millimeter-wave ultra massive MIMO systems,

    V . Jamali, A. M. Tulino, G. Fischer, R. R. M ¨uller, and R. Schober, “Intelligent surface-aided transmitter architectures for millimeter-wave ultra massive MIMO systems,” IEEE Open J. Commun. Soc. , vol. 2, pp. 144–167, 2021

    2021

  6. [14]

    How to deploy intelligent reflecting surfaces in wireless network: BS-side, user-side, or both sides?

    C. You, B. Zheng, W. Mei, and R. Zhang, “How to deploy intelligent reflecting surfaces in wireless network: BS-side, user-side, or both sides?” J. Commun. Inf. Netw. , vol. 7, no. 1, pp. 1–10, 2022

    2022

  7. [15]

    Integrating intelligent reflecting surface into base station: Architecture, channel model, and passive reflection design,

    Y . Huang, L. Zhu, and R. Zhang, “Integrating intelligent reflecting surface into base station: Architecture, channel model, and passive reflection design,” IEEE Trans. Commun., vol. 71, no. 8, pp. 5005–5020, 2023

    2023

  8. [16]

    Modeling and architecture design of reconfigurable intelligent surfaces using scattering parameter network analysis,

    S. Shen, B. Clerckx, and R. Murch, “Modeling and architecture design of reconfigurable intelligent surfaces using scattering parameter network analysis,” IEEE Trans. Wireless Commun., vol. 21, no. 2, pp. 1229–1243, 2022

    2022

  9. [17]

    Reconfigurable intelligent surfaces 2.0: Beyond diagonal phase shift matrices,

    H. Li, S. Shen, M. Nerini, and B. Clerckx, “Reconfigurable intelligent surfaces 2.0: Beyond diagonal phase shift matrices,” IEEE Commun. Mag., vol. 62, no. 3, pp. 102–108, 2024

    2024

  10. [18]

    Closed-form global optimization of beyond diagonal reconfigurable intelligent surfaces,

    M. Nerini, S. Shen, and B. Clerckx, “Closed-form global optimization of beyond diagonal reconfigurable intelligent surfaces,” IEEE Trans. Wireless Commun., vol. 23, no. 2, pp. 1037–1051, 2024

    2024

  11. [19]

    Stacked intelligent metasurfaces for efficient holographic MIMO communications in 6G,

    J. An, C. Xu, D. W. K. Ng, G. C. Alexandropoulos, C. Huang, C. Yuen, and L. Hanzo, “Stacked intelligent metasurfaces for efficient holographic MIMO communications in 6G,” IEEE J. Sel. Areas Commun. , vol. 41, no. 8, pp. 2380–2396, 2023

    2023

  12. [20]

    Two-dimensional direction-of-arrival estimation using stacked intelligent metasurfaces,

    J. An, C. Yuen, Y . L. Guan, M. D. Renzo, M. Debbah, H. V . Poor, and L. Hanzo, “Two-dimensional direction-of-arrival estimation using stacked intelligent metasurfaces,” IEEE J. Sel. Areas Commun. , vol. 42, no. 10, pp. 2786–2802, 2024

    2024

  13. [21]

    Physically consistent modeling of stacked intelligent metasurfaces implemented with beyond diagonal RIS,

    M. Nerini and B. Clerckx, “Physically consistent modeling of stacked intelligent metasurfaces implemented with beyond diagonal RIS,” IEEE Commun. Lett. , vol. 28, no. 7, pp. 1693–1697, 2024

    2024

  14. [22]

    Dynamic scattering arrays for simultaneous electromagnetic processing and radiation in holographic MIMO systems,

    D. Dardari, “Dynamic scattering arrays for simultaneous electromagnetic processing and radiation in holographic MIMO systems,” arXiv preprint arXiv:2405.16174, 2024

    2024 arXiv

  15. [23]

    Decentralized multi-antenna architectures with unitary constraints,

    J. V . Alegr ´ıa and O. Edfors, “Decentralized multi-antenna architectures with unitary constraints,” in 2024 58th Asilomar Conference on Signals, Systems, and Computers , 2024, pp. 1615–1619

    2024

  16. [24]

    Analog computing for signal processing and communications – Part I: Computing with microwave networks,

    M. Nerini and B. Clerckx, “Analog computing for signal processing and communications – Part I: Computing with microwave networks,” arXiv preprint arXiv:2504.06790, 2025

    2025

  17. [25]

    Analog computing for signal processing and communications – Part II: Toward gigantic MIMO beamforming,

    M. Nerini and B. Clerckx, “Analog computing for signal processing and communications – Part II: Toward gigantic MIMO beamforming,” arXiv preprint arXiv:2504.07477, 2025

    2025

  18. [26]

    Toward a circuit theory of communi- cation,

    M. T. Ivrla ˇc and J. A. Nossek, “Toward a circuit theory of communi- cation,” IEEE Trans. Circuits Syst. I: Reg. Papers , vol. 57, no. 7, pp. 1663–1683, 2010

    2010

  19. [27]

    A universal framework for multiport network analysis of reconfigurable intelligent surfaces,

    M. Nerini, S. Shen, H. Li, M. Di Renzo, and B. Clerckx, “A universal framework for multiport network analysis of reconfigurable intelligent surfaces,” IEEE Trans. Wireless Commun. , vol. 23, no. 10, pp. 14 575– 14 590, 2024

    2024

  20. [28]

    D. M. Pozar, Microwave engineering. John Wiley & Sons, 2011

    2011

  21. [29]

    T. M. Cover, Elements of information theory . John Wiley & Sons, 1999

    1999

  22. [30]

    Clerckx and C

    B. Clerckx and C. Oestges, MIMO wireless networks: Channels, techniques and standards for multi-antenna, multi-user and multi-cell systems. Academic Press, 2013

    2013

  23. [31]

    Completion of a symmetric unitary matrix,

    R. F. Mathis, “Completion of a symmetric unitary matrix,” SIAM Review, vol. 11, no. 2, pp. 261–263, 1969

    1969

  24. [32]

    Inverses of 2 ×2 block matrices,

    T.-T. Lu and S.-H. Shiou, “Inverses of 2 ×2 block matrices,” Computers & Mathematics with Applications , vol. 43, no. 1-2, pp. 119–129, 2002

    2002

  25. [33]

    Capacity of multi-antenna gaussian channels,

    E. Telatar, “Capacity of multi-antenna gaussian channels,” European Transactions on Telecommunications, vol. 10, no. 6, pp. 585–595, 1999

    1999

  26. [34]

    C. A. Balanis, Antenna theory: Analysis and design . John Wiley & Sons, 2015

    2015

Pith tools

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