pith. sign in

arxiv: 1906.11639 · v2 · pith:CRRRRUV4new · submitted 2019-06-26 · 💻 cs.IT · math.IT

On the Energy Efficiency of Limited-Backhaul Cell-Free Massive MIMO

Manijeh Bashar , Kanapathippillai Cumanan , Alister G. Burr , Hien Quoc Ngo , Erik G. Larsson , Pei Xiao This is my paper

Pith reviewed 2026-05-25 15:06 UTC · model grok-4.3

classification 💻 cs.IT math.IT
keywords cell-free massive MIMOenergy efficiencylimited backhaulquantizationsuccessive convex approximationBussgang theoremmaximum ratio combiningthroughput constraints
0
0 comments X

The pith

Sending quantized MRC signals over limited backhaul improves energy efficiency in cell-free massive MIMO

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

The paper examines energy efficiency maximization in cell-free Massive MIMO with limited backhaul capacity links. Access points use distributed maximum ratio combining so only quantized weighted signals are sent to the central processing unit. Channel estimation and quantization errors are modeled with the Bussgang theorem in the formulation of the optimization problem, which includes per-user power, backhaul capacity, and throughput constraints. The non-convex problem is decomposed into two sub-problems solved by successive convex approximation. Numerical results demonstrate the superiority of this approach.

Core claim

Thanks to the distributed maximum ratio combining weighting at the access points, only the quantized version of the weighted signals are sent back to the CPU. Considering the effects of channel estimation errors and using the Bussgang theorem to model the quantization errors, an energy efficiency maximization problem is formulated with per-user power and backhaul capacity constraints as well as with throughput requirement constraints. To handle this non-convex optimization problem, we decompose the original problem into two sub-problems and exploit a successive convex approximation to solve original energy efficiency maximization problem.

What carries the argument

Decomposition of the non-convex energy efficiency problem into two sub-problems solved via successive convex approximation, with quantization errors modeled by the Bussgang theorem.

If this is right

  • Numerical results confirm higher energy efficiency than other schemes.
  • The optimization meets all per-user power, backhaul capacity, and throughput constraints.
  • The approach accounts for channel estimation errors in addition to quantization effects.

Where Pith is reading between the lines

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

  • This method could lower the backhaul capacity needed for cell-free systems to operate efficiently.
  • Similar decomposition techniques might apply to other resource allocation problems in distributed MIMO.
  • Practical implementations could benefit from reduced infrastructure costs due to lower backhaul demands.

Load-bearing premise

The decomposition of the non-convex energy efficiency problem into two sub-problems combined with successive convex approximation produces a solution whose performance is representative of the original problem under the stated constraints.

What would settle it

Numerical experiments where an alternative method or exact solver yields a higher energy efficiency value under the same system parameters and constraints would falsify the claim of superiority.

Figures

Figures reproduced from arXiv: 1906.11639 by Alister G. Burr, Erik G. Larsson, Hien Quoc Ngo, Kanapathippillai Cumanan, Manijeh Bashar, Pei Xiao.

Figure 2
Figure 2. Figure 2: Average energy efficiency versus number of APs with [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The average energy efficiency of proposed Algorith [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The average energy efficiency of proposed Algorith [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
read the original abstract

We investigate the energy efficiency performance of cell-free Massive multiple-input multiple-output (MIMO), where the access points (APs) are connected to a central processing unit (CPU) via limited-capacity links. Thanks to the distributed maximum ratio combining (MRC) weighting at the APs, we propose that only the quantized version of the weighted signals are sent back to the CPU. Considering the effects of channel estimation errors and using the Bussgang theorem to model the quantization errors, an energy efficiency maximization problem is formulated with per-user power and backhaul capacity constraints as well as with throughput requirement constraints. To handle this non-convex optimization problem, we decompose the original problem into two sub-problems and exploit a successive convex approximation (SCA) to solve original energy efficiency maximization problem. Numerical results confirm the superiority of the proposed optimization scheme.

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 manuscript studies energy efficiency in limited-backhaul cell-free Massive MIMO. It proposes sending only quantized MRC-weighted signals from the APs to the CPU, models quantization noise via the Bussgang theorem, formulates a non-convex EE maximization problem subject to per-user power, backhaul capacity, and minimum-throughput constraints, decomposes the problem into two sub-problems, applies successive convex approximation (SCA) to obtain a solution, and reports numerical results that the proposed scheme outperforms baselines.

Significance. If the numerical gains are shown to be robust to the non-convexity, the work would provide a practical optimization approach for cell-free systems under realistic backhaul constraints. The modeling choices (distributed MRC + Bussgang) are standard and the decomposition idea is a reasonable attempt to handle the fractional objective.

major comments (2)
  1. [optimization formulation and SCA solution method] The central claim that numerical results confirm superiority rests on the output of the decomposed SCA procedure. No analysis is given showing that the obtained stationary point is close to the global optimum of the original non-convex problem, nor is an upper bound (e.g., via relaxation or exhaustive search on small instances) provided for comparison. This makes it impossible to rule out that the reported EE gains are artifacts of the particular local solution.
  2. [numerical results] The numerical-results section reports performance curves but does not state whether multiple random initializations were used for the SCA iterations or whether the same initialization strategy was applied to all compared schemes. Without this, the superiority claim cannot be assessed as representative of the original problem.
minor comments (2)
  1. [system model] Notation for the quantized signals and the Bussgang gain factor should be introduced with an explicit equation reference when first used.
  2. [abstract] The abstract should specify the simulation parameters (number of APs, users, backhaul capacities) that underlie the claimed superiority.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive comments. We address each major comment below and indicate the revisions we will make to strengthen the manuscript.

read point-by-point responses

  1. Referee: [optimization formulation and SCA solution method] The central claim that numerical results confirm superiority rests on the output of the decomposed SCA procedure. No analysis is given showing that the obtained stationary point is close to the global optimum of the original non-convex problem, nor is an upper bound (e.g., via relaxation or exhaustive search on small instances) provided for comparison. This makes it impossible to rule out that the reported EE gains are artifacts of the particular local solution.

    Authors: We agree that the manuscript does not provide a theoretical guarantee or upper bound establishing closeness to the global optimum. The SCA procedure yields a stationary point of the approximated problem, which is standard for this class of non-convex fractional programs but does not preclude the possibility of better local solutions. In the revised version we will add a dedicated paragraph in Section IV discussing the local nature of the solution, the convergence behavior of the algorithm, and the practical value of the obtained operating point. We will also include a small-scale exhaustive-search comparison on a reduced system to illustrate the gap to the global optimum where feasible. revision: yes

  2. Referee: [numerical results] The numerical-results section reports performance curves but does not state whether multiple random initializations were used for the SCA iterations or whether the same initialization strategy was applied to all compared schemes. Without this, the superiority claim cannot be assessed as representative of the original problem.

    Authors: The referee is correct that the current numerical section omits this information. In the revision we will explicitly state that (i) the same deterministic initialization (based on equal power allocation and uniform quantization) was used across all schemes for fairness, and (ii) additional Monte-Carlo trials with random initial points were performed; the best feasible objective value among those trials is reported for each scheme. The corresponding text and a short table summarizing the variation across initializations will be added to Section V. revision: yes

Circularity Check

0 steps flagged

No circularity in derivation chain

full rationale

The paper formulates an energy efficiency maximization problem for limited-backhaul cell-free Massive MIMO using the external Bussgang theorem to model quantization, decomposes the non-convex problem into two sub-problems, and applies successive convex approximation (SCA) to obtain a solution. Numerical results then evaluate the scheme's performance under the stated constraints. No load-bearing step reduces by construction to its own inputs via self-definition, fitted parameters renamed as predictions, or self-citation chains; the approach relies on standard external techniques and direct numerical evaluation rather than any renaming or forced equivalence. The derivation is therefore self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the applicability of the Bussgang theorem to model quantization and on the effectiveness of the SCA solver after decomposition; both are domain-standard assumptions rather than new entities or fitted constants introduced by the paper.

axioms (1)
  • domain assumption Bussgang theorem accurately models the quantization error when only the weighted signal is quantized at each AP
    Explicitly invoked in the abstract to handle quantization effects alongside channel estimation errors.

pith-pipeline@v0.9.0 · 5691 in / 1224 out tokens · 29037 ms · 2026-05-25T15:06:44.340770+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

33 extracted references · 33 canonical work pages

  1. [1]

    Cell-free Massive MIMO versus small cells,

    H. Q. Ngo, A. Ashikhmin, H. Y ang, E. G. Larsson, and T. L. Ma rzetta, “Cell-free Massive MIMO versus small cells,” IEEE Trans. Wireless Commun., vol. 16, no. 3, pp. 1834–1850, Mar. 2017

    work page 2017

  2. [2]

    On the total energy efficiency of cell-free massive MIMO,

    H. Q. Ngo, L. Tran, T. Q. Duong, M. Matthaiou, and E. G. Lars son, “On the total energy efficiency of cell-free massive MIMO,” IEEE Trans. Green Commun. and Net. , vol. 2, no. 1, pp. 25–39, Mar. 2017

    work page 2017

  3. [3]

    Enhanced max-min SINR for uplink cell-free Massive MIMO sy stems,

    M. Bashar, K. Cumanan, A. G. Burr, , M. Debbah, and H. Q. Ngo , “Enhanced max-min SINR for uplink cell-free Massive MIMO sy stems,” in Proc. IEEE ICC , May 2018, pp. 1–6

    work page 2018

  4. [4]

    On the uplink max-min SINR of cell-free massive MIMO systems,

    M. Bashar, K. Cumanan, A. G. Burr, M. Debbah, and H. Q. Ngo, “On the uplink max-min SINR of cell-free massive MIMO systems,” IEEE Trans. Wireless Commun. , pp. 1–17, Jan. 2019

    work page 2019

  5. [5]

    Ultra-dense radio ac cess networks for smart cities: Cloud-RAN, Fog-RAN and cell-free Massive MIMO,

    A. Burr, M. Bashar, and D. Maryopi, “Ultra-dense radio ac cess networks for smart cities: Cloud-RAN, Fog-RAN and cell-free Massive MIMO,” in Proc. IEEE PIMRC , Sep. 2018

    work page 2018

  6. [6]

    NOMA/OMA mode selection-based cell-free Massive MIMO,

    M. Bashar, K. Cumanan, A. Burr, H. Q. Ngo, L. Hanzo, and P . X iao, “NOMA/OMA mode selection-based cell-free Massive MIMO,” i n Proc. IEEE ICC , May 2019

    work page 2019

  7. [7]

    Cell- free Massive MIMO with limited backhaul,

    M. Bashar, K. Cumanan, A. G. Burr, H. Q. Ngo, and M. Debbah, “Cell- free Massive MIMO with limited backhaul,” in Proc. IEEE ICC , May 2018, pp. 1–7

    work page 2018

  8. [8]

    Max- min SINR of cell-free Massive MIMO uplink with optimal unifo rm quantization,

    M. Bashar, K. Cumanan, A. G. Burr, , H. Q. Ngo, and M. Debbah , “Max- min SINR of cell-free Massive MIMO uplink with optimal unifo rm quantization,” IEEE Trans. Commun. , submitted

    work page

  9. [9]

    Cooperative acces s networks: Optimum fronthaul quantization in distributed Massive MIM O and cloud RAN,

    A. G. Burr, M. Bashar, and D. Maryopi, “Cooperative acces s networks: Optimum fronthaul quantization in distributed Massive MIM O and cloud RAN,” in Proc. IEEE VTC , Jun. 2018, pp. 1–7

    work page 2018

  10. [10]

    On the performance of backhaul constrained cell- free Massive MIMO with linear receivers,

    M. Bashar, H. Q. Ngo, A. Burr, D. Maryopi, K. Cumanan, and E. G. Larsson, “On the performance of backhaul constrained cell- free Massive MIMO with linear receivers,” in Proc. IEEE Asilomar , Nov. 2018, pp. 1–7

    work page 2018

  11. [11]

    On the uplink throug hput of zero-forcing in cell-free Massive MIMO with coarse quantiz ation,

    D. Maryopi, M. Bashar, and A. Burr, “On the uplink throug hput of zero-forcing in cell-free Massive MIMO with coarse quantiz ation,” IEEE Trans. V eh. Technol., pp. 1–5, Accepted

    work page

  12. [12]

    Golub and C

    G. Golub and C. V . Loan, Matrix Computations , 2nd ed. Baltimore, MD: The Johns Hopkins Univ. Press, 1996

    work page 1996

  13. [13]

    Mixed quality of service in cell-free Massive MIMO,

    M. Bashar, K. Cumanan, A. G. Burr, H. Q. Ngo, and H. V . Poor , “Mixed quality of service in cell-free Massive MIMO,” IEEE Commun. Lett. , vol. 22, no. 7, pp. 706–709, Jul. 2018

    work page 2018

  14. [14]

    Joint beamforming and user maximization techniques for techniqu es for cognitive radio networks based on branch and bound method,

    K. Cumanan, R. Krishna, L. Musavian, and S. Lambotharan , “Joint beamforming and user maximization techniques for techniqu es for cognitive radio networks based on branch and bound method,” IEEE Trans. Wireless Commun. , vol. 9, no. 10, pp. 3082–3092, Oct. 2010

    work page 2010

  15. [15]

    MMSE based beamforming techniques for relay broadcast cha nnel,

    K. Cumanan, Y . Rahulamathavan, S. Lambotharan, and Z. D ing, “MMSE based beamforming techniques for relay broadcast cha nnel,” IEEE Trans. V eh. Technol., vol. 62, no. 8, pp. 4045–4051, Oct. 2013

    work page 2013

  16. [16]

    Rate balancin g based linear transceiver design for multiuser MIMO system with multiple linear transmit covariance constraints,

    K. Cumanan, J. Tang, and S. Lambotharan, “Rate balancin g based linear transceiver design for multiuser MIMO system with multiple linear transmit covariance constraints,” in Proc. IEEE ICC, Jun. 2011, pp. 1–5

    work page 2011

  17. [17]

    Bashar, Cell-free Massive MIMO and Millimeter W ave Channel Modelling for 5G and Beyond

    M. Bashar, Cell-free Massive MIMO and Millimeter W ave Channel Modelling for 5G and Beyond . Ph.D. dissertation, University of Y ork, United Kingdom, 2019

    work page 2019

  18. [18]

    Energy efficiency of the cell-free Massive MIMO upl ink with optimal uniform quantization,

    M. Bashar, K. Cumanan, A. G. Burr, H. Q. Ngo, E. G. Larsson , and P . Xiao, “Energy efficiency of the cell-free Massive MIMO upl ink with optimal uniform quantization,” IEEE Trans. Green Commun. and Net. , Accepted

    work page

  19. [19]

    Ener gy efficient coordinated beamforming for multicell system: duality-ba sed algorithm design and Massive MIMO transition,

    S. He, Y . Huang, L. Y ang, B. Ottersten, and W. Hong, “Ener gy efficient coordinated beamforming for multicell system: duality-ba sed algorithm design and Massive MIMO transition,” IEEE Trans. Commun. , vol. 63, no. 12, pp. 4920–4935, Dec. 2013

    work page 2013

  20. [20]

    Resource allocation for cross-layer utility maximizatio n in wireless networks,

    P . C. Weeraddana, M. Codreanu, M. Latva-aho, and A. Ephr emides, “Resource allocation for cross-layer utility maximizatio n in wireless networks,” IEEE Trans. V eh. Technol. , vol. 60, no. 6, pp. 2790–2809, Jul. 2011

    work page 2011

  21. [21]

    A study of dynamic multipath clusters at 60 GHz in a large indoor environment,

    M. Bashar, K. Haneda, A. Burr, and K. Cumanan, “A study of dynamic multipath clusters at 60 GHz in a large indoor environment,” in Proc. IEEE Globecom W orkshop, Dec. 2018

    work page 2018

  22. [22]

    Evaluation of low complexity Massi ve MIMO techniques under realistic channel conditions,

    M. Bashar, A. G. Burr, K. Haneda, K. Cumanan, M. M. Molu, M. Khalily, and P . Xiao, “Evaluation of low complexity Massi ve MIMO techniques under realistic channel conditions,” IEEE Trans. V eh. Technol., Accepted

    work page

  23. [23]

    Robust geometry-based user scheduling for large MIMO syst ems under realistic channel conditions,

    M. Bashar, A. G. Burr, D. Maryopi, K. Haneda, and K. Cuman an, “Robust geometry-based user scheduling for large MIMO syst ems under realistic channel conditions,” in Proc. IEEE EW , May 2018, pp. 1–6

    work page 2018

  24. [24]

    Relationship between two distortion mea sures for memo- ryless nonlinear systems,

    P . Zillmann, “Relationship between two distortion mea sures for memo- ryless nonlinear systems,” IEEE Signal Process. Lett. , vol. 17, no. 11, pp. 917–920, Feb. 2010

    work page 2010

  25. [25]

    A. V . Oppenheim, R. W. Schafer, and J. R. Buck, Discrete-time signal processing. Prentice-hall Englewood Cliffs, 1989

    work page 1989

  26. [26]

    The worst additive noise under a covariance con straint,

    J. Max, “The worst additive noise under a covariance con straint,” IRE Trans. Inf. Theory , vol. 6, no. 1, pp. 7–12, Nov. 1960. IEEE ICC, 2019 8

    work page 1960

  27. [27]

    Optimal design of energy-efficient multi-user MIMO systems: Is Massive MIM O the answer?

    E. Bj¨ ornson, L. Sanguinetti, J. Hoydis, and M. Debbah, “Optimal design of energy-efficient multi-user MIMO systems: Is Massive MIM O the answer?” IEEE Trans. Wireless Commun., vol. 14, no. 6, pp. 3059–3075, Jun. 2015

    work page 2015

  28. [28]

    Bit per joul e efficiency of cooperating base stations in cellular networks,

    A. J. Fehske, P . Marsch, and G. P . Fettweis, “Bit per joul e efficiency of cooperating base stations in cellular networks,” in Proc. IEEE Globecom W orkshops, Dec. 2010, pp. 1406–1411

    work page 2010

  29. [29]

    On the energy effi ciency- spectral efficiency trade-off of distributed MIMO systems,

    O. Onireti, F. Heliot, and M. A. Imran, “On the energy effi ciency- spectral efficiency trade-off of distributed MIMO systems, ” IEEE Trans. Commun., vol. 61, no. 9, pp. 3741–3753, Sep. 2013

    work page 2013

  30. [30]

    Towards energy efficiency with uplink cooperation in heterogeneous networks,

    L. Falconetti and E. Y assin, “Towards energy efficiency with uplink cooperation in heterogeneous networks,” in Proc. IEEE WCNC , Apr. 2014, pp. 1649–1654

    work page 2014

  31. [31]

    Coordinated beamforming for the m ulticell multi-antenna wireless system,

    H. Dahrouj and W. Y u, “Coordinated beamforming for the m ulticell multi-antenna wireless system,” IEEE Trans. Wireless Commun. , vol. 9, no. 5, pp. 1748–1759, Jan. 2010

    work page 2010

  32. [32]

    Weighted sum rate maximization for multiuser MISO systems with low resolution digital to analog converters,

    A. Kakkavas, J. Munir, A. Mezghani, H. Brunner, and J. A. Nossek, “Weighted sum rate maximization for multiuser MISO systems with low resolution digital to analog converters,” in Proc. IEEE WSA , Mar. 2016

    work page 2016

  33. [33]

    Boyd and L

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

    work page 2004