pith. sign in

arxiv: 2604.11471 · v1 · submitted 2026-04-13 · 📡 eess.SP

Stream-Adaptive Quantization and Power Allocation in Fronthaul-Constrained MIMO Systems

\"Ozlem Tu\u{g}fe Demir , Emil Bj\"ornson This is my paper

Pith reviewed 2026-05-10 15:18 UTC · model grok-4.3

classification 📡 eess.SP
keywords fronthaul quantizationMIMO systemsbit allocationpower allocationsum rate maximizationBussgang analysiscapacity lower boundstream-adaptive quantization
0
0 comments X

The pith

Joint bit and power allocation maximizes the sum rate in fronthaul-constrained MIMO systems, with uniform bit allocation becoming optimal at high SNR.

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

The paper extends a Bussgang-based analysis from quantized SISO channels to MIMO systems by performing receive combining before quantization to derive a capacity lower bound. It introduces a joint bit and power allocation scheme called JBP-Alloc to distribute fronthaul bits and transmit power across active data streams for sum rate maximization. Asymptotic analysis shows uniform bit allocation is optimal at high SNR, and numerical results demonstrate that this scheme outperforms uniform allocation and quantization-unaware water-filling while matching greedy allocation at lower complexity. A sympathetic reader would care because fronthaul capacity is a bottleneck in distributed wireless systems, and efficient allocation can improve performance without additional hardware.

Core claim

Starting from a Bussgang-based analysis of quantized SISO channels, the framework is extended to MIMO to derive a capacity lower bound under fronthaul quantization with receive combining before quantization. To maximize the sum rate, a joint bit and power allocation (JBP-Alloc) scheme efficiently distributes fronthaul bits and transmit power across active data streams. Asymptotic analysis shows that uniform bit allocation becomes optimal at high SNR.

What carries the argument

The JBP-Alloc scheme, which jointly optimizes bit allocation for fronthaul quantization and power allocation for transmission to maximize the sum rate in the derived MIMO capacity lower bound.

If this is right

  • JBP-Alloc achieves the same sum-rate performance as greedy bit allocation but with substantially lower computational complexity.
  • Uniform bit allocation becomes optimal at high SNR.
  • The scheme outperforms both uniform bit allocation and quantization-unaware water-filling power allocation.
  • Numerical results confirm these performance gains in various scenarios.

Where Pith is reading between the lines

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

  • The optimality of uniform allocation at high SNR could simplify system design in high-power regimes by eliminating the need for adaptive bit allocation.
  • This joint allocation approach may apply to other distributed antenna systems or cloud radio access networks with fronthaul constraints.
  • The capacity lower bound could be used as a benchmark for evaluating new quantization techniques in MIMO.

Load-bearing premise

The Bussgang-based analysis of quantized SISO channels extends to MIMO systems when receive combining is performed before quantization.

What would settle it

Simulations comparing the actual achievable sum rate with the derived lower bound under the JBP-Alloc scheme would falsify the claims if the bound is not tight or if the performance gains disappear.

Figures

Figures reproduced from arXiv: 2604.11471 by Emil Bj\"ornson, \"Ozlem Tu\u{g}fe Demir.

Figure 1
Figure 1. Figure 1: The sum rate in terms of the total number of [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
read the original abstract

Many wireless systems divide the baseband processing between two locations, interconnected by a fronthaul. This paper examines the impact of fronthaul quantization on multiple-input multiple-output (MIMO) systems. Starting from a Bussgang-based analysis of quantized single-input single-output (SISO) channels, we extend the framework to MIMO and derive a capacity lower bound under fronthaul quantization, where the receive combining is performed before the quantization. To maximize the sum rate, we propose a joint bit and power allocation (JBP-Alloc) scheme that efficiently distributes fronthaul bits and transmit power across active data streams. Asymptotic analysis shows that uniform bit allocation becomes optimal at high SNR. Numerical results confirm that JBP-Alloc outperforms uniform allocation and quantization-unaware water-filling, and achieves the same performance as Greedy bit allocation but with substantially lower computational complexity.

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 extends the Bussgang decomposition from quantized SISO channels to MIMO systems with linear receive combining performed before quantization. It derives a capacity lower bound under fronthaul quantization constraints, proposes a joint bit and power allocation (JBP-Alloc) scheme to maximize the sum rate by distributing fronthaul bits and transmit power across streams, shows via asymptotic analysis that uniform bit allocation is optimal at high SNR, and validates through numerical results that JBP-Alloc outperforms uniform allocation and quantization-unaware water-filling while achieving comparable performance to greedy allocation at substantially lower complexity.

Significance. If the central extension and lower bound hold, the work provides a practical, low-complexity resource allocation method for fronthaul-constrained MIMO systems such as C-RAN. The asymptotic high-SNR insight on uniform bit allocation is a useful design guideline, and the reported complexity reduction relative to greedy methods is a concrete strength. Numerical comparisons lend support to the performance claims under the stated model.

major comments (1)
  1. [Bussgang-based analysis and capacity lower bound derivation (following the SISO-to-MIMO extension)] The capacity lower bound (and thus the objective of the JBP-Alloc optimization) is obtained by extending the Bussgang additive quantization noise model from SISO to the MIMO case after receive combining. The manuscript does not re-derive or bound the cross-stream correlation terms that arise when the combining matrix is applied to signals with stream-dependent quantization bit allocations; the post-combining vector must satisfy the same uncorrelatedness and Gaussianity conditions as the scalar case for the noise variance to be used directly in the mutual-information expression. This assumption is load-bearing for the sum-rate formula and the subsequent allocation scheme.
minor comments (2)
  1. [Abstract] The abstract states that receive combining is performed before quantization but does not explicitly list the assumptions on channel state information and noise statistics that are used in the derivation.
  2. [System model and problem formulation] Notation for the quantization noise variance and the effective SNR after combining could be introduced earlier and used consistently in the optimization problem statement.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive and detailed review of our manuscript. We address the major comment on the Bussgang-based analysis and capacity lower bound derivation below.

read point-by-point responses

  1. Referee: [Bussgang-based analysis and capacity lower bound derivation (following the SISO-to-MIMO extension)] The capacity lower bound (and thus the objective of the JBP-Alloc optimization) is obtained by extending the Bussgang additive quantization noise model from SISO to the MIMO case after receive combining. The manuscript does not re-derive or bound the cross-stream correlation terms that arise when the combining matrix is applied to signals with stream-dependent quantization bit allocations; the post-combining vector must satisfy the same uncorrelatedness and Gaussianity conditions as the scalar case for the noise variance to be used directly in the mutual-information expression. This assumption is load-bearing for the sum-rate formula and the subsequent allocation scheme.

    Authors: We appreciate the referee highlighting this important modeling detail. The manuscript extends the scalar Bussgang decomposition component-wise to the post-combining vector, treating the quantization noise as additive and uncorrelated with the signal per stream, which directly yields the effective noise variance used in the mutual-information lower bound. While the cross-stream correlation terms arising from correlated post-combining signals are not explicitly bounded in the current text, we note that independent per-stream quantization implies the quantization noise vector has a covariance that can be expressed in terms of the Bussgang gains and the input correlation matrix. To address the concern rigorously, we will revise the manuscript by adding a dedicated derivation subsection that bounds the off-diagonal correlation terms (showing they are at most the product of the per-stream Bussgang factors times the normalized input correlations) and confirms that the diagonal approximation remains a valid lower bound. This addition will not alter the JBP-Alloc formulation, the asymptotic high-SNR result, or the numerical conclusions. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation extends Bussgang model and optimizes from the resulting bound

full rationale

The paper begins with the standard Bussgang decomposition for quantized SISO channels, performs an explicit extension to the MIMO case by placing linear receive combining before quantization, obtains a capacity lower bound from the resulting additive noise model, and then derives the JBP-Alloc scheme by optimizing that bound with respect to bit and power variables. The asymptotic claim that uniform bit allocation is optimal at high SNR follows directly from the closed-form behavior of the bound rather than from any fitted parameter or self-referential definition. No step reduces a claimed prediction to an input by construction, and the central optimization is not justified solely by self-citation; the numerical comparisons against uniform allocation and water-filling supply independent verification.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The work rests on standard wireless channel and quantization models; no new free parameters or invented entities are introduced in the abstract.

axioms (1)
  • domain assumption Bussgang theorem applies to the quantization model after receive combining in MIMO
    Used as the foundation for extending SISO analysis to MIMO capacity bound.

pith-pipeline@v0.9.0 · 5454 in / 1170 out tokens · 58701 ms · 2026-05-10T15:18:27.728422+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

14 extracted references · 14 canonical work pages

  1. [1]

    Capacity of multi-antenna Gaussian channe ls,

    E. Telatar, “Capacity of multi-antenna Gaussian channe ls,” European transactions on telecommunications , vol. 10, no. 6, pp. 585–595, 1999

    work page 1999

  2. [2]

    Optim ized precoding for MU-MIMO with fronthaul quantization,

    Y . Khorsandmanesh, E. Bj¨ ornson, and J. Jald´ en, “Optim ized precoding for MU-MIMO with fronthaul quantization,” IEEE Transactions on Wireless Communications, vol. 22, no. 11, pp. 7102–7115, 2023

    work page 2023

  3. [3]

    Capacity analysis of one-bit quant ized MIMO systems with transmitter channel state information,

    J. Mo and R. W. Heath, “Capacity analysis of one-bit quant ized MIMO systems with transmitter channel state information,” IEEE transactions on signal processing , vol. 63, no. 20, pp. 5498–5512, 2015. 6

    work page 2015

  4. [4]

    Quantized massive MU-MIMO-OFD M uplink,

    C. Studer and G. Durisi, “Quantized massive MU-MIMO-OFD M uplink,” IEEE Transactions on Communications , vol. 64, no. 6, pp. 2387–2399, 2016

    work page 2016

  5. [5]

    Capacity and coding for qu antized MIMO systems,

    J. A. Nossek and M. T. Ivrlaˇ c, “Capacity and coding for qu antized MIMO systems,” in Proceedings of the 2006 international conference on Wireless communications and mobile computing , 2006, pp. 1387–1392

    work page 2006

  6. [6]

    The Bussgang decomposition of nonlinear systems: Basic theory and MIMO extensions [lectu re notes],

    ¨O. T. Demir and E. Bj¨ ornson, “The Bussgang decomposition of nonlinear systems: Basic theory and MIMO extensions [lectu re notes],” IEEE Signal Processing Magazine , vol. 38, no. 1, pp. 131–136, 2020

    work page 2020

  7. [7]

    Robust predictive quantization: Analysis and design via convex op timization,

    A. K. Fletcher, S. Rangan, V . K. Goyal, and K. Ramchandran , “Robust predictive quantization: Analysis and design via convex op timization,” IEEE J. Sel. Topics Signal Process. , vol. 1, no. 4, pp. 618–632, 2007

    work page 2007

  8. [8]

    Gersho and R

    A. Gersho and R. M. Gray, V ector quantization and signal compression. Springer Science & Business Media, 2012, vol. 159

    work page 2012

  9. [9]

    Channel estimation and performance analysis of one-bit massive MIMO systems,

    Y . Li, C. Tao, G. Seco-Granados, A. Mezghani, A. L. Swindl ehurst, and L. Liu, “Channel estimation and performance analysis of one-bit massive MIMO systems,” IEEE Transactions on Signal Processing , vol. 65, no. 15, pp. 4075–4089, 2017

    work page 2017

  10. [10]

    How much training is need ed in multiple-antenna wireless links?

    B. Hassibi and B. M. Hochwald, “How much training is need ed in multiple-antenna wireless links?” IEEE Trans. Inf. Theory , vol. 49, no. 4, pp. 951–963, 2003

    work page 2003

  11. [11]

    Introduction to multiple antenna com- munications and reconfigurable surfaces,

    E. Bj¨ ornson and ¨O. T. Demir, “Introduction to multiple antenna com- munications and reconfigurable surfaces,” Now Publishers, Inc. , 2024

    work page 2024

  12. [12]

    Quantization,

    R. Gray and D. Neuhoff, “Quantization,” IEEE Transactions on Information Theory , vol. 44, no. 6, pp. 2325–2383, 1998

    work page 1998

  13. [13]

    Optimal resource alloc ation in coordi- nated multi-cell systems,

    E. Bj¨ ornson and E. Jorswieck, “Optimal resource alloc ation in coordi- nated multi-cell systems,” F oundations and Trends® in Communications and Information Theory , vol. 9, no. 2-3, pp. 113–381, 2013

    work page 2013

  14. [14]

    A. K. Jain, Fundamentals of digital image processing . Englewood Cliffs, NJ: Prentice Hall, 1989

    work page 1989