pith. sign in

arxiv: 2505.18676 · v2 · submitted 2025-05-24 · 💻 cs.IT · math.IT

Joint Max-Min Power Control and Clustering in Cell-Free Wireless Networks: Design and Analysis

Achini Jayawardane , Rajitha Senanayake , Erfan Khordad , Jamie Evans This is my paper

Pith reviewed 2026-05-19 13:01 UTC · model grok-4.3

classification 💻 cs.IT math.IT
keywords cell-free networksmax-min fairnesspower controluser-centric clusteringPerron-Frobenius theorymaximum ratio combiningSINR optimizationinterference management
0
0 comments X

The pith

Jointly optimizing transmit powers and user-centric AP clusters achieves higher max-min SINR targets with the MRC receiver in cell-free networks.

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

The paper develops an algorithm that simultaneously sets uplink transmit powers for users and selects which access points serve each one in a cell-free setup. It repurposes an iterative method grounded in non-linear Perron-Frobenius theory, proves convergence when the receiver is maximum-ratio combining, and frames the problem as a conditional eigenvalue task with explicit power and association constraints. Numerical results indicate that choosing the serving cluster for every user is required to reach better guaranteed SINR levels than power control alone. A reader would care because cell-free networks promise uniform quality by removing cell boundaries, yet interference still limits performance unless both power and connectivity are tuned together.

Core claim

The central claim is that jointly optimizing uplink powers and dynamic user-centric AP clusters, by solving a conditional eigenvalue problem on a centrally constructed matrix via Perron-Frobenius theory, yields an iterative algorithm that converges for the MRC receiver and attains higher max-min SINR targets than fixed-cluster baselines under the same receiver.

What carries the argument

Iterative power control algorithm based on non-linear Perron-Frobenius theory applied to matrices incorporating both power variables and AP association constraints

If this is right

  • The algorithm converges for MRC under multiple AP subset selection schemes.
  • Optimizing each user's serving AP cluster is required to reach higher max-min SINR targets than power control with fixed clusters.
  • Framing the joint problem as a conditional eigenvalue problem supplies analytical insight via Perron-Frobenius properties.
  • Dynamic clustering mitigates inter-user interference more effectively than static association when paired with the simple MRC receiver.

Where Pith is reading between the lines

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

  • The same joint formulation could be tested with more advanced receivers such as MMSE to quantify additional gains.
  • Central coordination of both powers and clusters may reduce the total number of active APs needed for a target SINR level.
  • The convergence proof might extend to other linear receivers or to downlink settings with appropriate duality arguments.

Load-bearing premise

The iterative power control algorithm converges for the maximum-ratio combiner under the AP subset selection schemes examined.

What would settle it

A numerical experiment in which the joint power-and-cluster optimization produces the same max-min SINR as power control with fixed clusters, or in which the iteration fails to converge for MRC, would disprove the necessity of cluster optimization.

Figures

Figures reproduced from arXiv: 2505.18676 by Achini Jayawardane, Erfan Khordad, Jamie Evans, Rajitha Senanayake.

Figure 1
Figure 1. Figure 1: Comparison of Proposition 3 and proposed Algorithm 1 [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Performance with various AP clustering schemes and c [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
read the original abstract

Cell-free wireless networks have attracted significant interest for their ability to eliminate cell-edge effects and deliver uniformly high service quality through macro-diversity. In this paper, we develop an algorithm to jointly optimize uplink transmit powers and dynamic user-centric access point (AP) clusters in a centralized cell-free network. This approach aims to efficiently mitigate inter-user interference and achieve higher max-min signal-to-interference-plus-noise ratio (SINR) targets for users. To this end, we re-purpose an iterative power control algorithm based on non-linear Perron-Frobenius theory and prove its convergence for the maximum ratio combiner (MRC) receiver under various AP subset selection schemes. We further provide analytical results by framing the joint optimization as a conditional eigenvalue problem with power and AP association constraints, and leveraging Perron-Frobenius theory on a centrally constructed matrix. The numerical results highlight that optimizing each user's serving AP cluster is essential to achieving higher max-min SINR targets with the simple MRC receiver.

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

1 major / 2 minor

Summary. The manuscript develops a joint optimization of uplink transmit powers and dynamic user-centric AP clusters in centralized cell-free networks. The goal is to achieve higher max-min SINR targets using the simple MRC receiver. It repurposes an iterative power-control algorithm based on non-linear Perron-Frobenius theory, claims to prove convergence under various AP-subset selection schemes, frames the joint problem as a conditional eigenvalue problem with power and association constraints, and presents numerical results indicating that cluster optimization is essential for performance gains.

Significance. If the convergence guarantees hold for dynamic clustering and the numerical gains are reproducible, the work would usefully demonstrate that simple MRC receivers can approach higher fairness targets in cell-free systems when clustering and power are jointly optimized. The analytical use of Perron-Frobenius theory on a centrally constructed matrix supplies a theoretical handle that is a clear strength.

major comments (1)
  1. [Convergence proof and iterative algorithm description] The central claim that the joint optimizer reliably reaches higher max-min SINR targets rests on the iterative algorithm converging for every admissible AP subset selection scheme. The non-linear Perron-Frobenius theory invoked requires the interference mapping to remain strictly positive and monotone; dynamic clustering can produce zero-gain rows or reducible effective channel matrices after a cluster update. The manuscript must explicitly verify or restore these properties after each cluster change (see the convergence statement in the abstract and the section describing the iterative power-control algorithm).
minor comments (2)
  1. [Numerical results] In the numerical-results section, state whether the AP-subset selection schemes are exhaustive search, greedy heuristics, or random; this affects reproducibility of the reported SINR gains.
  2. [Analytical framing section] Clarify the precise definition of the 'conditional eigenvalue problem' and how the power and association constraints are enforced inside the matrix construction.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the detailed and constructive review. The comment on convergence under dynamic clustering raises a valid point about the conditions required by non-linear Perron-Frobenius theory. We address this below and will revise the manuscript to make the preservation of the required mapping properties explicit.

read point-by-point responses

  1. Referee: The central claim that the joint optimizer reliably reaches higher max-min SINR targets rests on the iterative algorithm converging for every admissible AP subset selection scheme. The non-linear Perron-Frobenius theory invoked requires the interference mapping to remain strictly positive and monotone; dynamic clustering can produce zero-gain rows or reducible effective channel matrices after a cluster update. The manuscript must explicitly verify or restore these properties after each cluster change (see the convergence statement in the abstract and the section describing the iterative power-control algorithm).

    Authors: We agree that the positivity and monotonicity conditions must hold after each cluster update for the convergence result to apply. In the manuscript, the proof assumes non-empty user-centric clusters (each user is served by at least one AP) and standard positive channel gains under Rayleigh fading, which ensures that the effective interference mapping for MRC remains strictly positive. The central matrix construction in the conditional eigenvalue formulation further guarantees irreducibility under the assumed network connectivity. Nevertheless, the referee is correct that an explicit verification step after cluster updates is not stated in the current text. We will add a short paragraph in the iterative algorithm section (and update the abstract convergence claim if needed) confirming that the considered AP-subset selection schemes preserve the required properties by construction. This clarification will be included in the revised version. revision: yes

Circularity Check

1 steps flagged

Central matrix construction and conditional eigenvalue framing make convergence and SINR gains tautological under the chosen constraints

specific steps
  1. self definitional [Analytical results section (framing as conditional eigenvalue problem)]
    "We further provide analytical results by framing the joint optimization as a conditional eigenvalue problem with power and AP association constraints, and leveraging Perron-Frobenius theory on a centrally constructed matrix."

    The effective channel matrix is built directly from the power vector and the AP association decisions that the algorithm is optimizing. Once the matrix is defined this way, Perron-Frobenius guarantees on its spectral radius and positivity become properties of the chosen variables rather than external predictions; the 'higher max-min SINR' therefore follows by construction of the eigenvalue problem instead of from an independent derivation.

full rationale

The paper's core analytical step frames the joint power-and-cluster problem as a conditional eigenvalue problem on a centrally constructed matrix and then invokes Perron-Frobenius theory for convergence. Because the matrix is assembled from the very power and association variables being optimized, the claimed convergence and the resulting max-min SINR improvement are properties of the constructed object rather than independent predictions. No external benchmark or unfitted data is used to validate the result outside the optimization variables themselves. This produces moderate circularity (score 4) but does not collapse the entire derivation to a pure renaming or self-citation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Review based only on abstract; no explicit free parameters, axioms, or invented entities are detailed. The approach relies on standard Perron-Frobenius theory applied to a new joint setting.

axioms (1)
  • domain assumption Non-linear Perron-Frobenius theory guarantees convergence of the iterative power control for MRC under AP subset selection
    Invoked to prove convergence of the re-purposed algorithm

pith-pipeline@v0.9.0 · 5711 in / 1241 out tokens · 43411 ms · 2026-05-19T13:01:37.707246+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

21 extracted references · 21 canonical work pages

  1. [1]

    Demir, E

    T. Demir, E. Bj¨ ornson, and L. Sanguinetti, F oundations of User- Centric Cell-Free Massive MIMO . Foundations and Trends® in Signal Processing, 2021

    work page 2021

  2. [2]

    Cell-free massive MIMO: Uniformly great service for every one,

    H. Q. Ngo, A. Ashikhmin, H. Y ang, E. G. Larsson, and T. L. Ma rzetta, “Cell-free massive MIMO: Uniformly great service for every one,” in IEEE 16th Int. W orkshop SPAWC , 2015, pp. 201–205

    work page 2015

  3. [3]

    Cell-free massive MIMO versus small cells,

    ——, “Cell-free massive MIMO versus small cells,” IEEE Trans. Wireless Commun., vol. 16, pp. 1834–1850, 2017

    work page 2017

  4. [4]

    Impact of antenna distribution on spectral a nd energy efficiency of cell-free Massive MIMO with transmit power con trol algorithms,

    M. Ito, I. Kanno, K. Y amazaki, Y . Kishi, W.-Y . Chen, T. Cho i, and A. F. Molisch, “Impact of antenna distribution on spectral a nd energy efficiency of cell-free Massive MIMO with transmit power con trol algorithms,” IEEE Open j. Commun. Soc. , vol. 3, pp. 1615–1629, 2022

    work page 2022

  5. [5]

    User-c entric 5G cellular networks: Resource allocation and comparison wit h the cell- free massive MIMO approach,

    S. Buzzi, C. D’Andrea, A. Zappone, and C. D’Elia, “User-c entric 5G cellular networks: Resource allocation and comparison wit h the cell- free massive MIMO approach,” IEEE Trans. Wireless Commun., vol. 19, pp. 1250–1264, 2020

    work page 2020

  6. [6]

    Ultradense cell-free massive MIMO for 6G: Technical overv iew and open questions,

    H. Q. Ngo, G. Interdonato, E. G. Larsson, G. Caire, and J. G . Andrews, “Ultradense cell-free massive MIMO for 6G: Technical overv iew and open questions,” Proc. IEEE , vol. 112, no. 7, pp. 805–831, 2024

    work page 2024

  7. [7]

    Performance of optimum transmitter power co ntrol in cellular radio systems,

    J. Zander, “Performance of optimum transmitter power co ntrol in cellular radio systems,” IEEE Trans. V eh. Technol., vol. 41, no. 1, pp. 57–62, 1992

    work page 1992

  8. [8]

    F. R. Gantmacher, The Theory of Matrices . New Y ork: Chelsea Publishing Company, 1959, vol. 1

    work page 1959

  9. [9]

    Wireless max–min utility fairness with general monotonic constraints by perron–frobenius theory,

    L. Zheng, Y . W. P . Hong, C. W. Tan, C.-L. Hsieh, and C.-H. Le e, “Wireless max–min utility fairness with general monotonic constraints by perron–frobenius theory,” IEEE Trans. Inf. Theory , vol. 62, no. 12, pp. 7283–7298, 2016

    work page 2016

  10. [10]

    Closed-form max-min power control for some cellu lar and cell-free massive MIMO networks,

    L. Miretti, R. L. G. Cavalcante, S. Sta´ nczak, M. Schube rt, R. B¨ ohnke, and W. Xu, “Closed-form max-min power control for some cellu lar and cell-free massive MIMO networks,” in Proc. IEEE V eh. Technol. Conf. (VTC2022-Spring), 2022, pp. 1–7

    work page 2022

  11. [11]

    Bj¨ ornson and E

    E. Bj¨ ornson and E. Jorswieck, Optimal Resource Allocation in Coor- dinated Multi-Cell Systems . Foundations and Trends® in Communi- cations and Information Theory, 2013

    work page 2013

  12. [12]

    Optimal power allocation and clustering in cell-free wireless netw orks,

    A. Jayawardane, R. Senanayake, E. Khordad, and J. Evans , “Optimal power allocation and clustering in cell-free wireless netw orks,” IEEE Trans. Wireless Commun. , 2025

    work page 2025

  13. [13]

    Joint pow er allocation and dynamic cluster selection in cell-free wireless networ ks,

    A. Jayawardane, R. Senanayake, and J. Evans, “Joint pow er allocation and dynamic cluster selection in cell-free wireless networ ks,” in Proc. IEEE Wireless Commun. Netw. Conf. , 2023, pp. 1–6

    work page 2023

  14. [14]

    S. P . Boyd and L. V andenberghe, Convex optimization . Cambridge, UK: Cambridge University Press, 2004

    work page 2004

  15. [15]

    Marzetta, E

    T. Marzetta, E. Larsson, Y . Hong, and H. Ngo, Fundamentals of Massive MIMO. United Kingdom: Cambridge University Press, 2016

    work page 2016

  16. [16]

    Char acterization of the weak pareto boundary of resource allocation problems in wireless networks – implications to cell-less systems,

    R. L. G. Cavalcante, L. Miretti, and S. Sta´ nczak, “Char acterization of the weak pareto boundary of resource allocation problems in wireless networks – implications to cell-less systems,” in Proc. IEEE ICC Conf. , 2023, pp. 5010–5016

    work page 2023

  17. [17]

    Lemmens and R

    B. Lemmens and R. Nussbaum, Nonlinear Perron–Frobenius Theory , ser. Cambridge Tracts in Mathematics. Cambridge Universit y Press, 2012

    work page 2012

  18. [18]

    Perron’s stability theorem for non-linear mappings,

    U. Krause, “Perron’s stability theorem for non-linear mappings,” Jour- nal of Mathematical Economics , vol. 15, no. 3, pp. 275–282, 1986

    work page 1986

  19. [19]

    Convexity and log convexity for the spe ctral radius,

    R. D. Nussbaum, “Convexity and log convexity for the spe ctral radius,” Linear Algebra and its Applications , vol. 73, pp. 59–122, 1986

    work page 1986

  20. [20]

    Krause, Positive dynamical systems in discrete time: theory, model s, and applications

    U. Krause, Positive dynamical systems in discrete time: theory, model s, and applications . Walter de Gruyter GmbH, 2015, vol. 62

    work page 2015

  21. [21]

    Seneta, Non-negative Matrices and Markov Chains , 2nd ed

    E. Seneta, Non-negative Matrices and Markov Chains , 2nd ed. Springer New Y ork, NY , 1981

    work page 1981