pith. sign in

arxiv: 2506.15052 · v2 · submitted 2025-06-18 · 💻 cs.IT · eess.SP· math.IT

MIMO Systems Aided by Microwave Linear Analog Computers: Capacity-Achieving Architectures with Reduced Circuit Complexity

Matteo Nerini , Bruno Clerckx This is my paper

Pith reviewed 2026-05-19 09:46 UTC · model grok-4.3

classification 💻 cs.IT eess.SPmath.IT
keywords MIMOanalog beamformingmicrowave networkscircuit complexitycapacity-achievinggigantic MIMOgraph model
0
0 comments X

The pith

Stem-connected MiLACs achieve full MIMO capacity with circuit complexity that scales linearly in the number of antennas.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows that a new family of microwave linear analog computers, arranged in a stem-connected topology, can deliver the same performance as fully connected designs for MIMO beamforming. This matters for gigantic MIMO systems with thousands of antennas, where the number of required tunable components must stay manageable. The authors introduce a graph model of these networks to design the stem structure systematically and derive a closed-form solution that achieves capacity. Numerical results back up the analysis. A sympathetic reader sees a route to practical analog processing that avoids the quadratic hardware explosion of prior approaches.

Core claim

Stem-connected MiLACs are capacity-achieving for MIMO systems while requiring only a linear number of impedance components with the number of antennas, enabled by a graph theoretical model of the reconfigurable microwave networks and a closed-form optimization solution.

What carries the argument

The stem-connected MiLAC topology, defined via a graph model of microwave networks, which connects ports in a tree-like stem structure to realize capacity-achieving analog beamforming at linear complexity.

If this is right

  • Gigantic MIMO arrays become feasible without quadratic growth in analog components.
  • Closed-form solutions simplify the design of capacity-achieving beamformers.
  • Analog-domain processing can scale to larger antenna counts while keeping hardware practical.
  • High-performance MIMO systems for future networks gain a lower-complexity analog option.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar graph-based designs might extend to other reconfigurable analog circuits beyond microwave networks.
  • Power consumption in large arrays could drop because fewer tunable components are needed.
  • Real-world electromagnetic effects not captured by the graph model could limit performance gains.

Load-bearing premise

The graph model of the MiLAC accurately captures the electromagnetic behavior of the actual reconfigurable microwave networks.

What would settle it

Construct a physical stem-connected MiLAC prototype, measure its achieved rate over a real MIMO channel, and check whether it matches the theoretical capacity predicted by the model.

Figures

Figures reproduced from arXiv: 2506.15052 by Bruno Clerckx, Matteo Nerini.

Figure 1
Figure 1. Figure 1: MiLAC-aided MIMO system model. imaginary unit. IN and 0N denote the identity matrix and the all-zero matrix with dimensions N × N, respectively. 0M×N denotes the all-zero matrix with dimensions M × N. II. SYSTEM MODEL Consider a point-to-point MIMO system aided by a Mi￾LAC at both the transmitter and receiver, as introduced in [20, Section II] and represented in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Center graph with center size Q = 4. associated graph being a center graph as a stem-connected MiLAC, following the terminology introduced in BD-RIS literature [24], [25]. Based on this definition of center graph, the following proposition gives a sufficient condition for a transmitter-side MiLAC architecture to satisfy (16) for any V¯ , i.e., for being capacity-achieving. Proposition 1. A transmitter-side… view at source ↗
Figure 3
Figure 3. Figure 3: (a) Graph and (b) susceptance matrix (with tunable elements in gray) [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 5
Figure 5. Figure 5: Admittance matrix B of a stem-connected MiLAC at the transmitter partitioned into seven matrices, where B22,22 is diagonal. As a result, the problem of optimizing a stem-connected Mi￾LAC through a capacity-achieving solution can be formalized as a feasibility check problem, where we need to find a valid B partitioned as in (28) whose blocks satisfy the capacity￾achieving condition (29). Formally, we need t… view at source ↗
Figure 6
Figure 6. Figure 6: Admittance matrix B of a stem-connected MiLAC at the receiver partitioned into seven matrices, where B11,22 is diagonal. In summary, the problem of finding a capacity-achieving solution for a stem-connected MiLAC at the receiver side is find B11, B12, B21, B22 (78) s.t. B11 = B T 11, B21 = B T 12, B22 = B T 22, (79) B11 ∈ B11, (77) (80) JB11 = −Y0R, (81) JB12 = Y0INS , (82) RB11 + B21 = Y0J, (83) RB12 + B2… view at source ↗
Figure 8
Figure 8. Figure 8: Circuit complexity of stem- and fully-connected MiLAC versus the [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 7
Figure 7. Figure 7: Achievable rate versus the number of antennas [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
read the original abstract

To meet the demands of future wireless networks, antenna arrays must scale from massive multiple-input multiple-output (MIMO) to gigantic MIMO, involving even larger numbers of antennas. To address the hardware and computational cost of gigantic MIMO, several strategies are available that shift processing from the digital to the analog domain. Among them, microwave linear analog computers (MiLACs) offer a compelling solution by enabling fully analog beamforming through reconfigurable microwave networks. Prior work has focused on fully-connected MiLACs, whose ports are all interconnected to each other via tunable impedance components. Although such MiLACs are capacity-achieving, their circuit complexity, given by the number of required impedance components, scales quadratically with the number of antennas, limiting their practicality. To solve this issue, in this paper, we propose a graph theoretical model of MiLAC facilitating the systematic design of lower-complexity MiLAC architectures. Leveraging this model, we propose stem-connected MiLACs as a family of MiLAC architectures maintaining capacity-achieving performance while drastically reducing the circuit complexity. Besides, we optimize stem-connected MiLACs with a closed-form capacity-achieving solution. Our theoretical analysis, confirmed by numerical simulations, shows that stem-connected MiLACs are capacity-achieving, but with circuit complexity that scales linearly with the number of antennas, enabling high-performance, scalable, gigantic MIMO.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper proposes a graph-theoretic model for microwave linear analog computers (MiLACs) in MIMO systems. It introduces stem-connected MiLAC architectures that achieve MIMO channel capacity with circuit complexity scaling linearly (rather than quadratically) with the number of antennas. A closed-form capacity-achieving solution is derived, and the claims are supported by theoretical analysis plus numerical simulations.

Significance. If the graph model accurately captures the relevant electromagnetic transfer functions, the linear-complexity stem-connected design would represent a meaningful advance for scaling to gigantic MIMO, reducing hardware cost while preserving optimality. The closed-form solution is a positive feature, as it avoids iterative numerical optimization for the impedance settings.

major comments (2)
  1. [Graph model and closed-form solution sections] The capacity-achieving property is derived under the graph-theoretic abstraction of the MiLAC network (stem-connected topology). This abstraction must exactly reproduce the optimal precoding/combining matrices via port-to-port transfer functions. The manuscript does not provide a concrete argument or bound showing that sparsification to linear complexity preserves exact SVD matching when unmodeled effects (distributed parasitics, non-ideal tuner range) are present; see the modeling assumptions in the graph construction and the closed-form derivation.
  2. [Simulation results] Numerical results confirm performance only under the ideal graph model. No sensitivity analysis or Monte-Carlo study over model mismatch (e.g., impedance quantization or frequency-dependent S-parameters) is reported, which is load-bearing for the claim that the architecture remains capacity-achieving in a physical microwave implementation.
minor comments (2)
  1. [Architecture description] A diagram explicitly showing the stem-connected topology (nodes, stems, and tunable edges) would improve readability of the architectural description.
  2. [Abstract] The abstract states that complexity scales linearly but does not give the precise big-O expression or compare it directly to the fully-connected baseline in the same sentence.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments and the positive assessment of the paper's significance for scaling MIMO systems. We address each major comment below, clarifying the scope of our contributions and indicating planned revisions to improve clarity on modeling assumptions.

read point-by-point responses

  1. Referee: [Graph model and closed-form solution sections] The capacity-achieving property is derived under the graph-theoretic abstraction of the MiLAC network (stem-connected topology). This abstraction must exactly reproduce the optimal precoding/combining matrices via port-to-port transfer functions. The manuscript does not provide a concrete argument or bound showing that sparsification to linear complexity preserves exact SVD matching when unmodeled effects (distributed parasitics, non-ideal tuner range) are present; see the modeling assumptions in the graph construction and the closed-form derivation.

    Authors: The capacity-achieving property and closed-form solution are derived strictly within the graph-theoretic abstraction, where the stem-connected topology ensures that the port-to-port transfer functions exactly realize the optimal SVD-based precoding and combining matrices. This exact matching holds by construction in the model, as the graph structure and impedance assignments are chosen to replicate the required linear transformations with linear complexity. The manuscript does not claim or analyze performance under unmodeled physical effects such as distributed parasitics or non-ideal tuner ranges, which lie outside the graph abstraction. We will revise the graph construction and closed-form derivation sections to more explicitly restate the ideal modeling assumptions and add a dedicated paragraph in the conclusions discussing these limitations and identifying robustness to non-idealities as an important direction for future work. revision: partial

  2. Referee: [Simulation results] Numerical results confirm performance only under the ideal graph model. No sensitivity analysis or Monte-Carlo study over model mismatch (e.g., impedance quantization or frequency-dependent S-parameters) is reported, which is load-bearing for the claim that the architecture remains capacity-achieving in a physical microwave implementation.

    Authors: The numerical results are provided solely to validate the theoretical derivations and closed-form solution under the ideal graph model. They confirm that stem-connected MiLACs achieve the predicted MIMO capacity with the stated linear complexity. We agree that the lack of sensitivity analysis to mismatches such as impedance quantization or frequency-dependent S-parameters means the simulations do not directly address physical implementation robustness. We will revise the simulation results section to explicitly note that all results assume the ideal model and to include a qualitative discussion of how mismatches might affect performance based on the underlying graph-theoretic framework. A comprehensive Monte-Carlo study incorporating detailed electromagnetic models is beyond the current scope and is noted as future work. revision: partial

Circularity Check

0 steps flagged

New graph-theoretic MiLAC construction and closed-form optimization are self-contained

full rationale

The paper introduces a graph theoretical model of MiLAC networks, proposes stem-connected topologies, and derives a closed-form capacity-achieving solution for the network parameters. The capacity-achieving claim follows directly from solving the impedances under the model to realize the required linear transformations for MIMO capacity, supported by theoretical analysis and simulations. No derivation step reduces by the paper's equations to a fitted input renamed as prediction, a self-cited uniqueness result, or an ansatz smuggled from prior work; the central result is an independent architectural construction whose performance is verified within the stated model assumptions.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The paper introduces a graph-theoretic model of MiLACs and assumes standard MIMO capacity definitions hold for the analog beamformer; no explicit free parameters are named in the abstract, but the capacity-achieving claim rests on the new topology preserving the optimal singular vectors.

axioms (1)
  • domain assumption MIMO capacity is achieved when the analog beamforming network realizes the optimal singular vectors of the channel matrix.
    Implicit in the claim that the proposed architectures are capacity-achieving.
invented entities (1)
  • stem-connected MiLAC no independent evidence
    purpose: Lower-complexity capacity-achieving analog beamformer
    New family of architectures proposed in this work; no independent external evidence provided in abstract.

pith-pipeline@v0.9.0 · 5787 in / 1195 out tokens · 23233 ms · 2026-05-19T09:46:11.950214+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages

  1. [1]

    Five disruptive technology directions for 5G,

    F. Boccardi, R. W. Heath, A. Lozano, T. L. Marzetta, and P. Popovski, “Five disruptive technology directions for 5G,” IEEE Commun. Mag. , vol. 52, no. 2, pp. 74–80, 2014

    work page 2014

  2. [2]

    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

    work page 2014

  3. [3]

    (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

    work page 2022

  4. [4]

    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

    work page arXiv 2024

  5. [5]

    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

  6. [6]

    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

  7. [7]

    Reconfigurable antennas,

    R. L. Haupt and M. Lanagan, “Reconfigurable antennas,” IEEE Antennas Propag. Mag., vol. 55, no. 1, pp. 49–61, 2013

    work page 2013

  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

    work page arXiv 2025

  9. [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

    work page 2021

  10. [10]

    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

    work page 2023

  11. [11]

    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

    work page 2022

  12. [12]

    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

    work page 2024

  13. [13]

    Transmitter side beyond-diagonal reconfigurable intelligent surface for massive MIMO networks,

    A. Mishra, Y . Mao, C. D’Andrea, S. Buzzi, and B. Clerckx, “Transmitter side beyond-diagonal reconfigurable intelligent surface for massive MIMO networks,” IEEE Wireless Commun. Lett. , vol. 13, no. 2, pp. 352–356, 2024

    work page 2024

  14. [14]

    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

    work page 2023

  15. [15]

    Stacked intelligent metasurfaces for multiuser downlink beamforming in the wave domain,

    J. An, M. D. Renzo, M. Debbah, H. Vincent Poor, and C. Yuen, “Stacked intelligent metasurfaces for multiuser downlink beamforming in the wave domain,” IEEE Trans. Wireless Commun. , 2025

    work page 2025

  16. [16]

    Dynamic metasurface antennas for 6G extreme massive MIMO communications,

    N. Shlezinger, G. C. Alexandropoulos, M. F. Imani, Y . C. Eldar, and D. R. Smith, “Dynamic metasurface antennas for 6G extreme massive MIMO communications,” IEEE Wireless Commun. , vol. 28, no. 2, pp. 106–113, 2021

    work page 2021

  17. [17]

    Beyond-diagonal dynamic meta- surface antenna,

    H. Prod’homme and P. del Hougne, “Beyond-diagonal dynamic meta- surface antenna,” arXiv preprint arXiv:2504.13523 , 2025

    work page arXiv 2025

  18. [18]

    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

    work page arXiv 2025

  19. [19]

    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

    work page arXiv 2025

  20. [20]

    Capacity of MIMO systems aided by microwave linear analog computers (MiLACs),

    M. Nerini and B. Clerckx, “Capacity of MIMO systems aided by microwave linear analog computers (MiLACs),” arXiv preprint arXiv:2506.05983, 2025

    work page arXiv 2025

  21. [21]

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

    work page 2011

  22. [22]

    Beyond diagonal reconfig- urable intelligent surfaces utilizing graph theory: Modeling, architecture design, and optimization,

    M. Nerini, S. Shen, H. Li, and B. Clerckx, “Beyond diagonal reconfig- urable intelligent surfaces utilizing graph theory: Modeling, architecture design, and optimization,”IEEE Trans. Wireless Commun., vol. 23, no. 8, pp. 9972–9985, 2024

    work page 2024

  23. [23]

    J. A. Bondy and U. S. R. Murty, Graph theory with applications . Macmillan London, 1976, vol. 290

    work page 1976

  24. [24]

    A novel Q-stem connected architecture for beyond-diagonal reconfigurable intelligent surfaces,

    X. Zhou, T. Fang, and Y . Mao, “A novel Q-stem connected architecture for beyond-diagonal reconfigurable intelligent surfaces,” arXiv preprint arXiv:2411.18480, 2024

    work page arXiv 2024

  25. [25]

    Beyond diagonal RIS in multiuser MIMO: Graph theoretic modeling and optimal architectures with low complex- ity,

    Z. Wu and B. Clerckx, “Beyond diagonal RIS in multiuser MIMO: Graph theoretic modeling and optimal architectures with low complex- ity,” arXiv preprint arXiv:2502.16509 , 2025

    work page arXiv 2025

  26. [26]

    On the symmetric solutions of linear matrix equations,

    D. Hua, “On the symmetric solutions of linear matrix equations,” Linear Algebra and its Applications , vol. 131, pp. 1–7, 1990

    work page 1990