Pith. sign in

REVIEW 2 major objections 5 minor 40 references

Asymptotically Optimal Local Receiver in Uplink CF-mMIMO: A Functional-Variational Analysis

T0 review · 2 major / 5 minor · reviewed 2026-07-12 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read An uplink local receiver that optimizes the true ergodic rate needs no statistical LSFD at the CPU and still beats the standard LMMSE-LSFD design.

desk verdict Solid variational derivation of a simple local filter that drops LSFD and gains a few percent on true ergodic rate; load-bearing steps are asymptotic but the math is careful and the result is usable. read the letter →

arxiv 2607.02917 v1 pith:QKH2AJ53 submitted 2026-07-03 eess.SP

Jiafei Fu , Peng Jiang , Dongming Wang , Pengcheng Zhu This is my paper
classification eess.SP
keywords cell-freemassiveMIMOuplinklocalreceiverergodicratequasi-LMMSEfunctional-variationalanalysislarge-systemapproximationLSFDuse-and-then-forget
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

Cell-free massive MIMO systems usually detect uplink symbols with a local MMSE filter at each access point and then apply long-term statistical weights (LSFD) at the central processor. Those weights come from a conservative lower bound on the rate and require extra fronthaul and CPU computation. This paper instead designs the local filters directly for the true average rate under perfect local channel knowledge. Using a functional-variational argument and large-system random-matrix approximations, it obtains a closed-form quasi-LMMSE filter that points in the same direction as ordinary LMMSE but is scaled by an instantaneous channel-dependent scalar. Because that scalar already differs from AP to AP, a plain sum at the CPU is enough; statistical LSFD coefficients can be discarded entirely. Simulations show a consistent ergodic-rate gain (about 5 percent when each AP has few antennas) at strictly lower system complexity.

What carries the argument

Functional-variational stationarity under an expectation-based normalization constraint, together with a large-system argument that cross-AP coupling terms vanish in the conditional mean, which collapses the optimality condition to the closed-form Q-LMMSE receiver.

What would settle it

Measure true ergodic rates for Q-LMMSE (equal-gain sum) versus LMMSE-LSFD on a finite-size network (e.g., N=8 antennas per AP, M=20 APs) and check whether the claimed 5 percent gain disappears or reverses once the expectation-based constraint and vanishing-coupling approximations are no longer accurate.

Watch

Extended reading notes

Core claim

Under perfect local CSI, the asymptotically optimal local uplink receiver for the true ergodic rate is the quasi-LMMSE filter that inverts only the interference-plus-noise covariance (excluding the desired user) and multiplies by the desired channel. This filter shares the classical LMMSE direction up to an instantaneous CSI-dependent scalar; the scalar automatically supplies adaptive weighting when the local estimates are simply summed at the CPU, eliminating any need for statistical LSFD.

Load-bearing premise

The pointwise interference-plus-noise normalization can be replaced by its average, and all cross-AP interference terms average to zero given only local channel knowledge; both steps are claimed to become exact only when the total number of antennas is large.

Editorial extensions

If this is right

  • Local APs can keep ordinary LMMSE complexity while the CPU drops all LSFD matrix inversions and the associated statistical estimation overhead.
  • Fronthaul need only carry one scalar per user per AP; no long-term covariance statistics must be exchanged for combining.
  • When each AP has few antennas the ergodic-rate advantage is largest, exactly the regime where channel hardening is weakest and the UatF bound is loosest.
  • The same variational route also recovers classical LMMSE-LSFD as optimal under the UatF objective, giving a unified optimality proof for both metrics.

Reading between the lines

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

  • The same scalar-adaptive idea may transfer to other distributed detection problems where only local CSI is available and a central processor must fuse soft estimates without long-term statistics.
  • If the vanishing-coupling lemma can be sharpened to finite-N error bounds, one could certify how large a network must be before the closed-form Q-LMMSE is provably near-optimal.
  • Imperfect CSI and fronthaul quantization, listed as future work, are natural next tests: any residual gain under those impairments would strengthen the practical case for dropping LSFD.
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 / 5 minor

Summary. The paper designs uplink local receivers in cell-free massive MIMO directly from the true ergodic rate under perfect local CSI, rather than the conventional UatF lower bound. After replacing the pointwise interference-plus-noise normalization by an expectation-based surrogate and applying large-system random-matrix arguments, a functional-variational analysis yields the closed-form quasi-LMMSE (Q-LMMSE) receiver of Eq. (44). This filter shares the same direction as classical local LMMSE (Remark 2 / Sherman–Morrison) but differs by an instantaneous CSI-dependent scalar; the scalar varies across APs and thereby supplies adaptive weighting for equal-gain combining at the CPU, eliminating statistical LSFD coefficients and the associated CPU-side overhead. Appendix A separately proves that local MMSE plus LSFD is optimal under the UatF metric. Numerical comparisons under the true ergodic rate show a consistent gain over LMMSE-LSFD (approximately 5 % when N is small) at equal per-AP complexity.

Significance. If the asymptotic derivation is accepted, the work supplies a clean, closed-form local receiver that is optimal for the true ergodic-rate objective under distributed CSI constraints, while simultaneously removing the need for statistical LSFD. The functional-variational stationarity conditions, the fundamental lemma of complex-vector variational calculus, and the odd-symmetry argument that forces cross-AP coupling means to vanish (Appendix E) are carefully written and constitute a reusable technical contribution. The practical payoff—identical per-AP complexity to LMMSE yet strictly lower system-level complexity and a measurable ergodic-rate gain—is of direct interest to the CF-mMIMO community. The independent UatF-optimality proof in Appendix A is also valuable, as the literature has largely taken that optimality as given.

major comments (2)
  1. The collapse of the stationarity condition (34)–(37) to the claimed closed-form Q-LMMSE (44) rests on two successive large-system approximations whose accuracy is not quantified at the finite (M,N) used in the figures: (i) replacement of the pointwise interference-plus-noise constraint (23) by its expectation (26), asserted to be tight with relative error O(1/NM), and (ii) Lemma 2, which asserts that every cross-AP coupling term vanishes in the conditional mean given local CSI. Appendix E’s proof is elegant (circular symmetry + odd-function arguments after Sherman–Morrison + linear system whose only solution is zero), yet it is exact only under perfect independence or in the asymptotic limit. Residual coupling or residual fluctuation of the denominator at practical N=8–32, M=10–40 could shift the true stationary point of the original functional away from Eq. (44). A short numerical check
  2. Section IV-C.3 and Eqs. (42)–(44): after Lemma 2 removes the cross-AP terms, the remaining scalar η_kl(H_l) is still a high-dimensional random functional of the local channel. The paper asserts that it concentrates to a deterministic ¯η_k that can be absorbed by scale invariance. While concentration is standard, the manuscript never verifies how large the residual variation of η_kl remains for the simulated antenna counts, nor whether that residual variation itself contributes to the observed ergodic-rate gain. A brief Monte-Carlo plot of the coefficient of variation of η_kl versus N would clarify whether the practical receiver is truly the pure Q-LMMSE of (44) or a mildly channel-dependent variant.
minor comments (5)
  1. Abstract and Section V: the claimed “approximately 5 % gain when the number of antennas per AP is low” is visible in Fig. 2 but is never stated with an explicit numerical value or confidence interval in the text; a single sentence quantifying the gap at N=8 would help.
  2. Notation: the same symbol Γ_k is used for the noise-power diagonal in the UatF SINR (17) and for the aggregate desired-signal coefficient in Appendix B; a distinct symbol would avoid momentary confusion.
  3. Table I lists pilot power η_k even though the analysis assumes perfect CSI; either remove the unused parameter or clarify that it is retained only for possible future imperfect-CSI extensions.
  4. A few typographical slips remain (e.g., “Ergoic” in the heading of IV-A, “reveiver” in III-B, “V ariational” with a space). A careful proof-reading pass is needed.
  5. Fig. 1–4 captions and legends mix “Erg-Q-LMMSE”, “Erg-LMMSE-LSFD” and “UatF” curves; a short legend explanation in the text would improve readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Q-LMMSE form is obtained from a self-contained functional-variational stationarity analysis under explicit large-system relaxations, not by construction from its inputs or load-bearing self-citation.

full rationale

The derivation chain begins from the true ergodic-rate functional (22), introduces an explicit scale-invariance normalization (23), relaxes it to an expectation surrogate (26) justified by channel hardening / RMT concentration of order O(1/NM), forms the Lagrangian (28), computes functional derivatives (31)–(33), invokes the fundamental lemma of variational calculus (Lemma 1) to obtain the conditional stationarity condition (34), decomposes the cross-AP terms, and then proves via circular-symmetry / odd-function / Sherman–Morrison arguments (Appendix E) that those terms vanish in conditional expectation (Lemma 2). The resulting local filter (41) further concentrates to the deterministic-scalar Q-LMMSE (44). None of these steps defines the receiver in terms of the claimed rate, fits a free parameter to data and re-labels it a prediction, or imports a uniqueness theorem from overlapping authors that forbids alternatives. Appendix A independently recovers classical LMMSE+LSFD under the UatF objective by the same variational machinery; that is a parallel result, not a circular premise. Self-citations ([4],[30],[31]) supply background CF-mMIMO context and are not load-bearing for the identity (44). The numerical ~5 % gain is a Monte-Carlo evaluation of the derived closed form against LMMSE-LSFD, not a fitted quantity. Approximations (expectation constraint, vanishing coupling) may be imperfect at finite (M,N), but that is a correctness / tightness issue, not circularity by construction. The paper is therefore self-contained against the circularity patterns enumerated.

Assumptions & free parameters 0 free parameters · 6 assumptions · 1 invented entities

The central claim rests on standard wireless and random-matrix background plus three modeling choices that are not theorems: perfect local CSI, replacement of the pointwise SINR normalization by an expectation constraint, and the large-system vanishing of cross-AP conditional means. No numerical constants are fitted to produce the claimed receiver; simulation parameters only evaluate it.

assumptions (6)
  • domain assumption Perfect local CSI at each AP (h_km known exactly at AP m).
    Stated in §II and used throughout the functional optimization; imperfect CSI is deferred to future work.
  • domain assumption Channels independent across APs and UEs; h_km ~ CN(0, R_km).
    §II system model; independence is essential for the conditional-expectation decoupling in Lemma 2 / Appendix E.
  • ad hoc to paper Pointwise interference-plus-noise constraint may be replaced by its expectation (Eq. 26) with relative error O(1/NM) in the large-system limit.
    §IV-B.2; this surrogate makes the variational problem tractable and is not proved for the finite (M,N) used in simulations.
  • ad hoc to paper Cross-AP coupling terms have conditional mean zero given local CSI (Lemma 2).
    Proved under Gaussian circular symmetry and independence in Appendix E; load-bearing for full AP decoupling.
  • standard math Scale invariance of instantaneous SINR allows normalization of the denominator to 1 without loss of optimality.
    §IV-B.1; standard for Rayleigh-quotient / SINR problems.
  • standard math Fundamental lemma of complex-vector-valued variational calculus (Lemma 1).
    Appendix C; used to pass from vanishing real-part inner products to almost-everywhere stationarity.
invented entities (1)
  • Q-LMMSE (quasi-LMMSE) local receiver independent evidence
    purpose: Closed-form local combiner claimed to be asymptotically optimal for the true ergodic rate under distributed CSI, enabling LSFD-free CPU fusion.
    Derived object rather than a free postulate; still the named construct the paper introduces and evaluates. Independent evidence is the implementable formula and the Monte-Carlo rate comparisons.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Asymptotically Optimal Local Receiver in Uplink CF-mMIMO: A Functional-Variational Analysis." pith.science (2026). https://pith.science/paper/QKH2AJ53

@misc{pith2026260702917,
  author       = {Pith},
  title        = {Pith review of: Asymptotically Optimal Local Receiver in Uplink CF-mMIMO: A Functional-Variational Analysis},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QKH2AJ53}},
  note         = {Machine review of arXiv:2607.02917}
}
read the original abstract

In cell-free massive multiple-input multiple-output (CF-mMIMO) systems, the canonical uplink local receiver is the local minimum mean square error (LMMSE) receiver with large-scale fading decoding (LSFD) at the central processing unit (CPU). The LSFD coefficients are derived under the use-and-then-forget (UatF) lower bound of the ergodic rate, and computing these coefficients introduces additional fronthaul overhead and computational complexity at the CPU. This paper investigates local receiver design directly from the true ergodic-rate objective under perfect local channel state information (CSI). By introducing an expectation-based constraint and leveraging large-system random matrix theory, we develop a functional-variational approach that yields the asymptotically optimal quasi-LMMSE (Q-LMMSE) receiver in closed form. A key insight is that the Q-LMMSE receiver shares the same direction as the conventional LMMSE receiver, differing only by an instantaneous CSI-dependent scalar, and thus incurs the same per-access point (AP) complexity. More importantly, this scalar varies across APs and implicitly provides adaptive weighting for the direct summation at the CPU, thereby completely eliminating the need for statistical LSFD coefficients and the associated CPU-side computational overhead. Numerical results demonstrate that the proposed Q-LMMSE receiver consistently outperforms the LMMSE-LSFD benchmark in terms of the ergodic rate, achieving approximately a {5\%} gain when the number of antennas per AP is low, while operating with strictly lower system-level complexity.

Figures

Figures reproduced from arXiv: 2607.02917 by the authors.

Figure 1
Figure 1. Average ergodic and UatF rates versus the number of AP [PITH_FULL_IMAGE:figures/full_fig_p008_1.png] view at source ↗
Figure 2
Figure 2. Average ergodic and UatF rates versus the number of an [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Average ergodic and UatF rates versus the number of us [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

40 extracted references

  1. [1]

    The road to 6G: Ten physical layer ch allenges for communications engineers,

    M. Matthaiou, O. Y urduseven, H. Q. Ngo, D. Morales-Jimen ez, S. L. Cotton, and V . F. Fusco, “The road to 6G: Ten physical layer ch allenges for communications engineers,” IEEE Commun. Mag. , vol. 59, no. 1, pp. 64–69, Feb. 2021

    2021

  2. [2]

    Towards 6G wireless communication networks: vision, enabling technologies, and new paradigm shifts,

    X. Y ou, C. Wang, J. Huang et al., “Towards 6G wireless communication networks: vision, enabling technologies, and new paradigm shifts,” Sci. China Inf. Sci. , vol. 64, no. 1, Jan. 2020

    2020

  3. [3]

    On the road to 6G: Visions, requirements, key technologies, and testbeds,

    C.-X. Wang et al. , “On the road to 6G: Visions, requirements, key technologies, and testbeds,” IEEE Commun. Surveys Tuts., vol. 25, no. 2, pp. 905–974, Feb. 2023

    2023

  4. [4]

    Optim al local precoding for cell-free massive MIMO under the UatF bound,

    P . Jiang, J. Fu, P . Zhu, D. Wang, Y . Wang, and X. Y ou, “Optim al local precoding for cell-free massive MIMO under the UatF bound,” IEEE Trans. Signal Process. , vol. 74, no. 1, pp. 2224–2239, 2026

    2026

  5. [5]

    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, Jan. 2017

    2017

  6. [6]

    User-centric cell-free massive MIMO networks: A survey of opportunities, challenges and solutions,

    H. A. Ammar, R. Adve, S. Shahbazpanahi, G. Boudreau, and K . V . Srinivas, “User-centric cell-free massive MIMO networks: A survey of opportunities, challenges and solutions,” IEEE Commun. Surveys Tuts. , vol. 24, no. 1, pp. 611–652, Dec. 2022

    2022

  7. [7]

    Cell-f ree massive MIMO in O-RAN: Energy-aware joint orchestration of cloud, fronthaul, and radio resources,

    ¨O. T. Demir, M. Masoudi, E. Bj¨ ornson, and C. Cavdar, “Cell-f ree massive MIMO in O-RAN: Energy-aware joint orchestration of cloud, fronthaul, and radio resources,” IEEE J. Sel. Areas Commun. , vol. 42, no. 2, pp. 356–372, Jan. 2024

    2024

  8. [8]

    Spectral efficiency of cell-free massive MIMO with LSFD receiver under spatially correlated rayleigh fad ing channels,

    L. Ning and F. Pingzhi, “Spectral efficiency of cell-free massive MIMO with LSFD receiver under spatially correlated rayleigh fad ing channels,” China Commun. , vol. 22, no. 8, pp. 1–18, Sep. 2025

    2025

Show all 40 references
  1. [9]

    In terde- pendent cell-free and cellular networks: Thinking the role of cell-free architecture for 6G,

    Y . Zhang, Q. Hu, M. Peng, Y . Liu, G. Zhang, and T. Jiang, “In terde- pendent cell-free and cellular networks: Thinking the role of cell-free architecture for 6G,” IEEE Netw. , vol. 38, no. 5, pp. 247–254, Nov. 2024

    2024

  2. [10]

    Cell-free mas sive MIMO: A survey,

    S. Elhoushy, M. Ibrahim, and W. Hamouda, “Cell-free mas sive MIMO: A survey,” IEEE Commun. Surveys Tuts. , vol. 24, no. 1, pp. 492–523, Oct. 2022

    2022

  3. [11]

    Local parti al zero- forcing combining for cell-free massive MIMO systems,

    J. Zhang, J. Zhang, E. Bj¨ ornson, and B. Ai, “Local parti al zero- forcing combining for cell-free massive MIMO systems,” IEEE Trans. Commun., vol. 69, no. 12, pp. 8459–8473, Dec. 2021

    2021

  4. [12]

    Performance of cell-f ree massive MIMO with rician fading and phase shifts,

    ¨O. ¨Ozdogan, E. Bj¨ ornson, and J. Zhang, “Performance of cell-f ree massive MIMO with rician fading and phase shifts,” IEEE Trans. Wireless Commun., vol. 18, no. 11, pp. 5299–5315, Aug. 2019

    2019

  5. [13]

    Bj¨ ornson, J

    E. Bj¨ ornson, J. Hoydis, and L. Sanguinetti, F oundations of User-Centric Cell-Free Massive MIMO . Now Publishers, 2021

    2021

  6. [14]

    On joint local comb ining for large-scale-fading decoding aided user-centric cell-fre e massive MIMO,

    Y . Zhang, S. Li, X. Zhu, and T. Jiang, “On joint local comb ining for large-scale-fading decoding aided user-centric cell-fre e massive MIMO,” IEEE Trans. V eh. Technol., vol. 75, no. 1, pp. 1671–1676, Jan. 2026

    2026

  7. [15]

    Perf ormance analysis for user-centric cell-free massive MIMO systems w ith hardware impairments and multi-antenna users,

    M. Xie, X. Y u, Y . Rui, K. Wang, X. Dang, and J. Zhang, “Perf ormance analysis for user-centric cell-free massive MIMO systems w ith hardware impairments and multi-antenna users,” IEEE Trans. Wireless Commun. , vol. 23, no. 2, pp. 1243–1259, Feb. 2024

    2024

  8. [16]

    Hardware-impaired rician-faded cell-free massive MIMO s ystems with LSFD and channel aging: SE analysis and optimization,

    V . Tentu, D. N. Amudala, A. Chattopadhyay, and R. Budhir aja, “Hardware-impaired rician-faded cell-free massive MIMO s ystems with LSFD and channel aging: SE analysis and optimization,” IEEE Trans. Commun., vol. 71, no. 9, pp. 5455–5472, May 2023

    2023

  9. [17]

    Uplink precoding design for cell-free massive MIMO with iteratively weighte d MMSE,

    Z. Wang, J. Zhang, H. Q. Ngo, B. Ai, and M. Debbah, “Uplink precoding design for cell-free massive MIMO with iteratively weighte d MMSE,” IEEE Trans. Commun. , vol. 71, no. 3, pp. 1532–1547, Jan. 2023

    2023

  10. [18]

    Optimal b ilinear equalizer beamforming design for cell-free massive MIMO ne tworks with arbitrary channel estimators,

    Z. Wang, J. Zhang, H. Lei, D. Niyato, and B. Ai, “Optimal b ilinear equalizer beamforming design for cell-free massive MIMO ne tworks with arbitrary channel estimators,” IEEE Trans. V eh. Technol., vol. 74, no. 3, pp. 1–6, Dec. 2025

    2025

  11. [19]

    Low-complexity distributed combining design for near-fie ld cell-free XL-MIMO systems,

    Z. Wang, J. Zhang, B. Xu, D. Niyato, B. Ai, S. Mao, and Z. Ha n, “Low-complexity distributed combining design for near-fie ld cell-free XL-MIMO systems,” IEEE Trans. Wireless Commun. , vol. 25, no. 1, pp. 11 799–11 815, Feb. 2026

    2026

  12. [20]

    RIS-assisted cell-free massive MIMO relying on reflection pattern modulation,

    Z. Sui, H. Q. Ngo, T. V an Chien, M. Matthaiou, and L. Hanzo , “RIS-assisted cell-free massive MIMO relying on reflection pattern modulation,” IEEE Trans. Commun. , vol. 73, no. 2, pp. 968–982, Aug. 2024

    2024

  13. [21]

    Uplink performan ce of RIS-aided cell-free massive MIMO system with electromag netic interference,

    E. Shi, J. Zhang, D. W. K. Ng, and B. Ai, “Uplink performan ce of RIS-aided cell-free massive MIMO system with electromag netic interference,” IEEE J. Sel. Areas Commun. , vol. 41, no. 8, pp. 2431– 2445, Jun. 2023

    2023

  14. [22]

    Joint sen sing and bi-directional communication with dynamic TDD enabled cel l-free MIMO,

    A. Chowdhury, S. S. Thoota, and E. G. Larsson, “Joint sen sing and bi-directional communication with dynamic TDD enabled cel l-free MIMO,” IEEE Trans. Wireless Commun., vol. 25, no. 1, pp. 2420–2434, Aug. 2026

    2026

  15. [23]

    Uplink resource allocation optimiza tion for user- centric cell-free MIMO networks,

    Z. Li and R. Adve, “Uplink resource allocation optimiza tion for user- centric cell-free MIMO networks,” IEEE Trans. Wireless Commun. , vol. 23, no. 10, pp. 12 622–12 637, May 2024

    2024

  16. [24]

    Massive M IMO has unlimited capacity,

    E. Bj¨ ornson, J. Hoydis, and L. Sanguinetti, “Massive M IMO has unlimited capacity,” IEEE Trans. Inf. Theory , vol. 17, no. 1, pp. 574– 590, Nov. 2018

    2018

  17. [25]

    Joint power control and LSFD fo r wireless-powered cell-free massive MIMO,

    ¨O. T. Demir and E. Bj¨ ornson, “Joint power control and LSFD fo r wireless-powered cell-free massive MIMO,” IEEE Trans. Wireless Com- mun., vol. 20, no. 3, pp. 1756–1769, Nov. 2021

    2021

  18. [26]

    Scalable cell-free m assive MIMO systems,

    E. Bj¨ ornson and L. Sanguinetti, “Scalable cell-free m assive MIMO systems,” IEEE Trans. Commun. , vol. 68, no. 7, pp. 4247–4261, Apr. 2020

    2020

  19. [27]

    Couillet and M

    R. Couillet and M. Debbah, Random Matrix Methods for Wireless Communications. Cambridge University Press, 2011

    2011

  20. [28]

    Sanguinetti, E

    L. Sanguinetti, E. Bj¨ ornson, and J. Hoydis, Random Matrix Theory for Wireless Communications. Now Publishers, 2022

    2022

  21. [29]

    Larg e system analysis of cooperative multi-cell downlink transmission via regularized channel inversion with imperfect CSIT,

    J. Zhang, C.-K. Wen, S. Jin, X. Gao, and K.-K. Wong, “Larg e system analysis of cooperative multi-cell downlink transmission via regularized channel inversion with imperfect CSIT,” IEEE Trans. Wireless Commun., vol. 12, no. 10, pp. 4801–4813, Aug. 2013

    2013

  22. [30]

    Local partial RZF in cell-free massive MIMO: A deterministic equi valent analysis,

    J. Fu, P . Jiang, P . Zhu, H. Q. Ngo, M. Matthaiou, and X. Y ou , “Local partial RZF in cell-free massive MIMO: A deterministic equi valent analysis,” IEEE Trans. Wireless Commun. , vol. 25, no. 1, pp. 12 548– 12 564, Feb. 2026

    2026

  23. [31]

    Perf ormance analysis of local partial MMSE precoding-based user-centr ic cell-free massive MIMO systems and deployment optimization,

    P . Jiang, J. Fu, P . Zhu, Y . Wang, J. Wang, and X. Y ou, “Perf ormance analysis of local partial MMSE precoding-based user-centr ic cell-free massive MIMO systems and deployment optimization,” IEEE Trans. Wireless Commun., vol. 25, no. 1, pp. 5701–5717, Oct. 2026

    2026

  24. [32]

    Massive MIMO in t he UL/DL of cellular networks with many antennas,

    J. Hoydis, S. ten Brink, and M. Debbah, “Massive MIMO in t he UL/DL of cellular networks with many antennas,” IEEE J. Sel. Areas Commun. , vol. 31, no. 2, pp. 160–171, Aug. 2013

    2013

  25. [33]

    La rge system analysis of linear precoding in correlated MISO broadcast c hannels under limited feedback,

    S. Wagner, R. Couillet, M. Debbah, and D. T. M. Slock, “La rge system analysis of linear precoding in correlated MISO broadcast c hannels under limited feedback,” IEEE Trans. Inf. Theory , vol. 58, no. 7, pp. 4509–4537, Mar. 2012

    2012

  26. [34]

    Noncooperative cellular wireless wit h unlimited num- bers of base station antennas,

    T. L. Marzetta, “Noncooperative cellular wireless wit h unlimited num- bers of base station antennas,” IEEE Trans. Wireless Commun. , vol. 9, no. 11, pp. 3590–3600, Oct. 2010

    2010

  27. [35]

    Ener gy efficiency optimization for downlink massive MIMO with stat istical CSIT,

    L. Y ou, J. Xiong, X. Yi, J. Wang, W. Wang, and X. Gao, “Ener gy efficiency optimization for downlink massive MIMO with stat istical CSIT,” IEEE Trans. Wireless Commun. , vol. 19, no. 4, pp. 2684–2698, Jan. 2020

    2020

  28. [36]

    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, Feb. 2015

    2015

  29. [37]

    Statis tical 3-D beamforming for large-scale MIMO downlink systems over ric ian fading channels,

    X. Li, S. Jin, H. A. Suraweera, J. Hou, and X. Gao, “Statis tical 3-D beamforming for large-scale MIMO downlink systems over ric ian fading channels,” IEEE Trans. Commun. , vol. 64, no. 4, pp. 1529–1543, Feb. 2016

    2016

  30. [38]

    Spatial channel model for multiple i nput multiple output MIMO simulations,

    E. T. S. I. (ETSI), “Spatial channel model for multiple i nput multiple output MIMO simulations,” ETSI Technical Report TR 125 996, Sept. 2020, corresponds to 3GPP TR 25.996 version 16.0.0 Release 1 6

    2020

  31. [39]

    ¨Erl¨ auterungen zu den Anfangsgr¨ unden der V ari- ationsrechnung,

    P . du Bois-Raymond, “ ¨Erl¨ auterungen zu den Anfangsgr¨ unden der V ari- ationsrechnung,” Math. Ann. , vol. 15, no. 1, pp. 283–314, Jun. 1879

  32. [40]

    R. A. Horn and C. R. Johnson, Matrix Analysis , 2nd ed. Cambridge University Press, 2013

    2013

Pith tools

Reviewed July 12, 2026 · model on record in the stance chip above.