pith. sign in

arxiv: 2605.15950 · v1 · pith:R3YVECNInew · submitted 2026-05-15 · 💻 cs.IT · math.IT

Vectorized Generalized Nearest Neighbor Decoding for In-block Memory Channel

Yuhao Liu , Xinwei Li , Shuqin Pang , Hao Wu , Wenyi Zhang This is my paper

Pith reviewed 2026-05-19 19:28 UTC · model grok-4.3

classification 💻 cs.IT math.IT
keywords vectorized generalized nearest neighbor decodingin-block memory channelsgeneralized mutual informationmismatch capacityGaussian codebooksphase noise channelsnoncoherent communication
0
0 comments X

The pith

For in-block memory channels the optimal vectorized generalized nearest neighbor decoder admits an analytical characterization when Gaussian codebooks are employed.

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

The paper extends generalized nearest neighbor decoding from memoryless settings to a vectorized form that processes blocks of symbols together to handle memory confined inside each block. Using the generalized mutual information as a practical lower bound on mismatch capacity, it derives closed-form expressions for the best decoding metrics under Gaussian inputs and states the conditions under which those metrics are optimal. A joint optimization of transmit covariance and receiver metric is formulated that reduces to a covariance search once the metric is solved in closed form; for diagonal covariances the resulting optimality conditions are self-consistent equations. The method is evaluated on block noncoherent additive white Gaussian noise and phase noise channels, where it outperforms conventional scaling baselines. Readers care because many practical wireless links exhibit precisely this form of intra-block memory, and a receiver that can be tuned analytically offers a direct route to higher reliability without transmitter changes.

Core claim

Leveraging the generalized mutual information as an operational lower bound on the mismatch capacity, an analytical characterization of the optimal Vec-GNND is obtained for general IBM channels with Gaussian codebooks. The formalism further provides closed-form optimality conditions and achievable GMIs for restricted variants of the receiver architecture. A GMI-based joint design viewpoint for Gaussian codebook covariance and decoding metrics is formulated; since the metric optimization admits a closed-form solution for each fixed covariance, the joint design reduces to an input-covariance optimization problem, and first-order self-consistent optimality conditions are derived for the family.

What carries the argument

The vectorized generalized nearest neighbor decoder (Vec-GNND), a block-wise extension of the generalized distance metric that incorporates in-block memory by operating on vectors of received symbols rather than scalar symbols.

If this is right

  • For any fixed input covariance the optimal decoding metric admits a closed-form solution.
  • The joint covariance-metric design therefore reduces to an input-covariance optimization problem.
  • For the diagonal covariance family first-order self-consistent optimality conditions can be written explicitly.
  • On block noncoherent AWGN and phase noise channels the resulting receiver yields measurable rate gains over scaling-based baselines.

Where Pith is reading between the lines

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

  • The same GMI-based metric derivation may be applied to channels whose memory spans multiple blocks if the vector length is increased accordingly.
  • The closed-form metric expressions could be used inside adaptive algorithms that track slow changes in channel statistics without retraining the entire receiver.
  • Because the method separates covariance choice from metric choice, it may combine with existing constellation optimization techniques that already assume Gaussian-like second-order statistics.

Load-bearing premise

That Gaussian codebooks are used and that the generalized mutual information lower bound remains tight enough to identify the truly optimal vectorized receiver for arbitrary in-block memory channels.

What would settle it

On a concrete IBM channel, compute the GMI achieved by the analytically derived Vec-GNND metric and compare it to the GMI obtained by numerically maximizing the metric over all possible forms; if the analytical form is not maximal then the characterization does not hold.

Figures

Figures reproduced from arXiv: 2605.15950 by Hao Wu, Shuqin Pang, Wenyi Zhang, Xinwei Li, Yuhao Liu.

Figure 1
Figure 1. Figure 1: Illustration of an in-block memory channel. The message [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: GMI comparison for element-wise GNNDR and optimal Vec-GNNDR with block length for block noncoherent AWGN channel of [PITH_FULL_IMAGE:figures/full_fig_p020_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Performance comparison of the optimal (unrestricted) Vec-GNNDR, restricted optimal Vec-GNNDR variants, and the identity decoding rule. The [PITH_FULL_IMAGE:figures/full_fig_p021_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Comparison of achievable GMI for the optimal (unrestricted) Vec-GNNDR, its restricted variants, and the identity decoding rule. The results are [PITH_FULL_IMAGE:figures/full_fig_p022_4.png] view at source ↗
read the original abstract

This work extends the generalized nearest neighbor decoding (GNND), originally developed as a receiver architecture for memoryless channels, to a vectorized GNND (Vec-GNND) suitable for in-block memory (IBM) channels. Leveraging the generalized mutual information (GMI) as an operational lower bound on the mismatch capacity, an analytical characterization of the optimal Vec-GNND is obtained for general IBM channels with Gaussian codebooks. The formalism further provides closed-form optimality conditions and achievable GMIs for restricted variants of the receiver architecture. Furthermore, we formulate a GMI-based joint design viewpoint for Gaussian codebook covariance and decoding metrics. Since the metric optimization admits a closed-form solution for each fixed covariance, the joint design is reduced to an input-covariance optimization problem; for the diagonal covariance family, we derive first-order self-consistent optimality conditions. Numerical evaluations on block noncoherent additive white Gaussian noise channels and phase noise channels demonstrate consistent performance gains over conventional scaling-based baselines, highlighting the substantial advantages and potential relevance of the proposed Vec-GNND in realistic communication scenarios.

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 extends generalized nearest neighbor decoding (GNND) to a vectorized form (Vec-GNND) for in-block memory (IBM) channels. Using the generalized mutual information (GMI) as an operational lower bound on mismatch capacity, it derives an analytical characterization of the optimal Vec-GNND for general IBM channels under Gaussian codebooks, along with closed-form optimality conditions for restricted receiver variants. A joint design of input covariance and decoding metrics is formulated, reducing to covariance optimization with first-order self-consistent conditions derived for the diagonal covariance family. Numerical results on block noncoherent AWGN and phase noise channels report performance gains over scaling-based baselines.

Significance. If the GMI remains a sufficiently tight proxy for the mismatch capacity on general IBM channels, the closed-form conditions and joint-design reduction would provide a practical analytical tool for receiver optimization in channels with memory. The reduction of metric optimization to a covariance problem for fixed metrics is a clean structural contribution, and the numerical gains over conventional baselines indicate potential relevance for realistic scenarios. The work would benefit from explicit verification that the derived conditions optimize actual performance rather than the GMI proxy alone.

major comments (2)
  1. [Abstract] Abstract (paragraph 2): The analytical characterization of the 'optimal Vec-GNND' treats GMI as an operational lower bound sufficient to support optimality conclusions and closed-form conditions for arbitrary IBM transition kernels. No gap bounds, tightness conditions, or general verification steps are stated; the numerical comparisons are confined to two specific channel families (block noncoherent AWGN, phase noise) where tightness may hold coincidentally, leaving the central claim dependent on an unverified proxy.
  2. [Joint design] Joint design section (self-consistent optimality conditions): The first-order optimality conditions for the diagonal covariance family are derived from the same GMI objective used to characterize the Vec-GNND. This creates a potential circular dependence between the claimed prediction of optimality and the fitted covariance, which must be resolved to confirm that the conditions are not tautological.
minor comments (2)
  1. [Abstract] The abstract refers to 'consistent performance gains' and 'substantial advantages' without detailing the exact baseline implementations, exclusion rules for comparisons, or error-analysis methodology; adding these would improve reproducibility.
  2. Notation for the vectorized metric and the restricted receiver architectures could be introduced with a short table or diagram to clarify the distinctions among variants.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments and the opportunity to clarify the scope of our results. We address each major comment below, qualifying our claims with respect to the GMI lower bound and explaining the derivation of the optimality conditions. Proposed revisions focus on explicit statements to avoid overclaiming.

read point-by-point responses

  1. Referee: [Abstract] Abstract (paragraph 2): The analytical characterization of the 'optimal Vec-GNND' treats GMI as an operational lower bound sufficient to support optimality conclusions and closed-form conditions for arbitrary IBM transition kernels. No gap bounds, tightness conditions, or general verification steps are stated; the numerical comparisons are confined to two specific channel families (block noncoherent AWGN, phase noise) where tightness may hold coincidentally, leaving the central claim dependent on an unverified proxy.

    Authors: We agree that the results characterize Vec-GNND optimality with respect to the GMI, which serves as a lower bound on mismatch capacity rather than equaling it for arbitrary kernels. The closed-form conditions and analytical characterization are derived under the GMI objective for general IBM channels with Gaussian codebooks. We will revise the abstract and introduction to explicitly state that 'optimal' refers to GMI maximization and add a remark noting the absence of general tightness bounds or gap characterizations, as these depend on specific transition kernels and are left for future work. The numerical evaluations on the two channel families demonstrate practical gains under the GMI metric; we will also add a sentence acknowledging that tightness may vary by channel. This revision clarifies the proxy nature without altering the technical contributions. revision: partial

  2. Referee: [Joint design] Joint design section (self-consistent optimality conditions): The first-order optimality conditions for the diagonal covariance family are derived from the same GMI objective used to characterize the Vec-GNND. This creates a potential circular dependence between the claimed prediction of optimality and the fitted covariance, which must be resolved to confirm that the conditions are not tautological.

    Authors: The self-consistent conditions arise after first optimizing the metric in closed form for any fixed covariance (yielding the GMI expression as a function of covariance alone) and then setting the gradient of this reduced GMI to zero with respect to the diagonal covariance entries. This is a standard reduction in joint optimization and is not circular or tautological; the conditions identify stationary points of the covariance optimization problem under the optimal metric. We will insert a clarifying paragraph in the joint design section that walks through this two-step derivation and states that the conditions are necessary for local optimality of the reduced problem. This should address the concern directly. revision: yes

Circularity Check

0 steps flagged

No circularity: GMI-based optimization is standard and self-contained

full rationale

The paper treats GMI as an external operational lower bound on mismatch capacity (a standard information-theoretic quantity) and derives the Vec-GNND characterization, closed-form metric solutions, and first-order self-consistent optimality conditions for diagonal covariance by direct optimization of this objective. The joint design reduces covariance optimization to a standard problem after closed-form metric step, without any self-definition, fitted-input-as-prediction, or load-bearing self-citation. No equations or claims reduce the target result to its own inputs by construction; the derivation remains independent of the specific IBM kernels and is validated numerically on concrete channels. This is the normal case of a self-contained analysis.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the standard information-theoretic assumption that GMI lower-bounds mismatch capacity and on the modeling choice of Gaussian codebooks; no new entities are postulated.

axioms (1)
  • domain assumption GMI serves as an operational lower bound on the mismatch capacity
    Explicitly leveraged in the abstract to obtain the analytical characterization of optimal Vec-GNND.

pith-pipeline@v0.9.0 · 5722 in / 1240 out tokens · 62453 ms · 2026-05-19T19:28:17.182306+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.

  • IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    Leveraging the generalized mutual information (GMI) as an operational lower bound on the mismatch capacity, an analytical characterization of the optimal Vec-GNND is obtained for general IBM channels with Gaussian codebooks.

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

49 extracted references · 49 canonical work pages

  1. [1]

    A mathematical theory of communication,

    C. E. Shannon, “A mathematical theory of communication,”The Bell system technical journal, vol. 27, no. 3, pp. 379–423, Jul. 1948

    work page 1948

  2. [2]

    I. M. Jacobs and J. Wozencraft,Principles of communication engineering.New York, NY , USA: Wiley, 1965

    work page 1965

  3. [3]

    Massive MIMO systems with non-ideal hardware: Energy efficiency, estimation, and capacity limits,

    E. Bj ¨ornson, J. Hoydis, M. Kountouris, and M. Debbah, “Massive MIMO systems with non-ideal hardware: Energy efficiency, estimation, and capacity limits,”IEEE Transactions on information theory, vol. 60, no. 11, pp. 7112–7139, Nov. 2014

    work page 2014

  4. [4]

    Reliable communication under channel uncertainty,

    A. Lapidoth and P. Narayan, “Reliable communication under channel uncertainty,”IEEE transactions on Information Theory, vol. 44, no. 6, pp. 2148– 2177, Nov. 2002

    work page 2002

  5. [5]

    Mismatched decoding: Error exponents, second-order rates and saddlepoint approximations,

    J. Scarlett, A. Martinez, and A. G. i F `abregas, “Mismatched decoding: Error exponents, second-order rates and saddlepoint approximations,”IEEE Transactions on Information Theory, vol. 60, no. 5, pp. 2647–2666, May 2014

    work page 2014

  6. [6]

    On information rates for mismatched decoders,

    N. Merhav, G. Kaplan, A. Lapidoth, and S. S. Shitz, “On information rates for mismatched decoders,”IEEE Transactions on Information Theory, vol. 40, no. 6, pp. 1953–1967, Nov. 1994

    work page 1953

  7. [7]

    Channel capacity for a given decoding metric,

    I. Csiszar and P. Narayan, “Channel capacity for a given decoding metric,”IEEE Transactions on Information Theory, vol. 41, no. 1, pp. 35–43, Jan. 2002

    work page 2002

  8. [8]

    Information-theoretic foundations of mismatched decoding,

    J. Scarlett, A. G. i F `abregas, A. Somekh-Baruch, A. Martinezet al., “Information-theoretic foundations of mismatched decoding,”Foundations and Trends® in Communications and Information Theory, vol. 17, no. 2–3, pp. 149–401, 2020

    work page 2020

  9. [9]

    Some remarks on the role of inaccuracy in Shannon’s theory of information transmission,

    T. R. Fischer, “Some remarks on the role of inaccuracy in Shannon’s theory of information transmission,” inProc. 8th Prague Conference on Information Theory, Statistical Decision Functions, Random Processes. Prague, Czechoslovakia: Springer, Aug. 1978, pp. 211–226, volume A, held from August 28 to September 1, 1978

    work page 1978

  10. [10]

    On information rates of compound channels,

    G. Kaplan and S. Shamai, “On information rates of compound channels,” inProc. IEEE International Symposium on Information Theory (ISIT), Budapest, Hungary, 1991, p. 100

    work page 1991

  11. [11]

    Achievable information rates for fiber optics: Applications and computations,

    A. Alvarado, T. Fehenberger, B. Chen, and F. M. J. Willems, “Achievable information rates for fiber optics: Applications and computations,”Journal of Lightwave Technology, vol. 36, no. 2, pp. 424–439, Jan. 2018

    work page 2018

  12. [12]

    Constellation design with geometric and probabilistic shaping,

    S. Zhang and F. Yaman, “Constellation design with geometric and probabilistic shaping,”Optics Communications, vol. 402, pp. 7–12, Nov. 2017

    work page 2017

  13. [13]

    End-to-end learning for GMI-optimized geometric constellation shape,

    R. T. Jones, M. P. Yankov, and D. Zibar, “End-to-end learning for GMI-optimized geometric constellation shape,” inProc. 45th European Conference on Optical Communication (ECOC), Dublin, Ireland, 2019, pp. 1–4

    work page 2019

  14. [14]

    GMI-maximizing constellation design with Grassmann projection for parametric shaping,

    T. Koike-Akino, D. S. Millar, K. Parsons, and K. Kojima, “GMI-maximizing constellation design with Grassmann projection for parametric shaping,” inProc. Optical Fiber Communication Conference and Exhibition (OFC). Anaheim, CA, USA: Optica Publishing Group, Mar. 2016, pp. M2A–4

    work page 2016

  15. [15]

    A double maximization approach for optimizing the LM rate of mismatched decoding,

    L. Chen, S. Wu, X. Li, H. Wu, H. Wu, and W. Zhang, “A double maximization approach for optimizing the LM rate of mismatched decoding,” inProc. IEEE International Symposium on Information Theory (ISIT). Athens, Greece: IEEE, 2024, pp. 1504–1509

    work page 2024

  16. [16]

    A general framework for transmission with transceiver distortion and some applications,

    W. Zhang, “A general framework for transmission with transceiver distortion and some applications,”IEEE Transactions on Communications, vol. 60, no. 2, pp. 384–399, Feb. 2011

    work page 2011

  17. [17]

    A remark on channels with transceiver distortion,

    W. Zhang, “A remark on channels with transceiver distortion,” inProc. Information Theory and Applications Workshop (ITA). La Jolla, CA, USA: IEEE, Jan. 2016, pp. 1–4

    work page 2016

  18. [18]

    A regression approach to certain information transmission problems,

    W. Zhang, Y . Wang, C. Shen, and N. Liang, “A regression approach to certain information transmission problems,”IEEE Journal on Selected Areas in Communications, vol. 37, no. 11, pp. 2517–2531, Nov. 2019

    work page 2019

  19. [19]

    Crosscorrelation functions of amplitude-distorted Gaussian signals,

    J. J. Bussgang, “Crosscorrelation functions of amplitude-distorted Gaussian signals,” Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, MA, USA, Tech. Rep., 1952

    work page 1952

  20. [20]

    Generalized nearest neighbor decoding,

    Y . Wang and W. Zhang, “Generalized nearest neighbor decoding,”IEEE Transactions on Information Theory, vol. 68, no. 9, pp. 5852–5865, Sep. 2022

    work page 2022

  21. [21]

    Generalized nearest neighbor decoding: General input constellation and a case study of interference suppression,

    S. Pang and W. Zhang, “Generalized nearest neighbor decoding: General input constellation and a case study of interference suppression,”IEEE Transactions on Communications, vol. 73, no. 1, pp. 1–15, Jan. 2025

    work page 2025

  22. [22]

    Generalized nearest neighbor decoding for MIMO channels with imperfect channel state information,

    S. Pang and W. Zhang, “Generalized nearest neighbor decoding for MIMO channels with imperfect channel state information,” inProc. IEEE Information Theory Workshop (ITW). Kanazawa, Japan: IEEE, Oct. 2021, pp. 1–6

    work page 2021

  23. [23]

    Linear shrinkage receiver for slow fading channels under imperfect channel state information,

    W. Shi, S. Pang, and W. Zhang, “Linear shrinkage receiver for slow fading channels under imperfect channel state information,” inProc. IEEE Information Theory Workshop (ITW). Mumbai, India: IEEE, Nov. 2022, pp. 338–343

    work page 2022

  24. [24]

    J. G. Proakis and M. Salehi,Digital communications, 4th ed. McGraw-hill New York, 2001

    work page 2001

  25. [25]

    Fading channels: Information-theoretic and communications aspects,

    E. Biglieri, J. Proakis, and S. Shamai, “Fading channels: Information-theoretic and communications aspects,”IEEE transactions on information theory, vol. 44, no. 6, pp. 2619–2692, Nov. 2002

    work page 2002

  26. [26]

    Models and information rates for Wiener phase noise channels,

    H. Ghozlan and G. Kramer, “Models and information rates for Wiener phase noise channels,”IEEE Transactions on Information Theory, vol. 63, no. 4, pp. 2376–2393, Apr. 2017

    work page 2017

  27. [27]

    On the capacity of channels with block memory,

    W. Stark and R. McEliece, “On the capacity of channels with block memory,”IEEE Transactions on Information Theory, vol. 34, no. 2, pp. 322–324, Mar. 1988

    work page 1988

  28. [28]

    Information networks with in-block memory,

    G. Kramer, “Information networks with in-block memory,”IEEE Transactions on Information Theory, vol. 60, no. 4, pp. 2105–2120, Apr. 2014

    work page 2014

  29. [29]

    On phase noise channels at high SNR,

    A. Lapidoth, “On phase noise channels at high SNR,” inProc. IEEE Information Theory Workshop (ITW). Bangalore, India: IEEE, Oct. 2002, pp. 1–4

    work page 2002

  30. [30]

    Capacity and coding for the block-independent noncoherent AWGN channel,

    R. Nuriyev and A. Anastasopoulos, “Capacity and coding for the block-independent noncoherent AWGN channel,”IEEE transactions on information theory, vol. 51, no. 3, pp. 866–883, Mar. 2005

    work page 2005

  31. [31]

    Compensation of phase noise in OFDM wireless systems,

    Q. Zou, A. Tarighat, and A. H. Sayed, “Compensation of phase noise in OFDM wireless systems,”IEEE transactions on signal processing, vol. 55, no. 11, pp. 5407–5424, Nov. 2007

    work page 2007

  32. [32]

    Achievable rates in Gaussian MISO channels with secrecy constraints,

    S. Shafiee and S. Ulukus, “Achievable rates in Gaussian MISO channels with secrecy constraints,” inProc. IEEE International Symposium on Information Theory (ISIT). Nice, France: IEEE, Jun. 2007, pp. 2466–2470

    work page 2007

  33. [33]

    Secrecy capacity of SIMO and slow fading channels,

    P. Parada and R. Blahut, “Secrecy capacity of SIMO and slow fading channels,” inProc. International Symposium on Information Theory (ISIT). Adelaide, Australia: IEEE, Sep. 2005, pp. 2152–2155

    work page 2005

  34. [34]

    On fractionally-spaced equalizer design for digital microwave radio channels,

    C. Johnson, H. Lee, J. LeBlanc, T. Endres, R. Casas, E. Tai, Z. Reznic, and W. Meyer, “On fractionally-spaced equalizer design for digital microwave radio channels,” inProc. 29th Asilomar Conference on Signals, Systems and Computers, vol. 1. Pacific Grove, CA, USA: IEEE, Oct. 1995, pp. 290–294

    work page 1995

  35. [35]

    Data transmission by frequency-division multiplexing using the discrete Fourier transform,

    S. Weinstein and P. Ebert, “Data transmission by frequency-division multiplexing using the discrete Fourier transform,”IEEE transactions on Communication Technology, vol. 19, no. 5, pp. 628–634, Oct. 1971

    work page 1971

  36. [36]

    Wireless multicarrier communications,

    Z. Wang and G. B. Giannakis, “Wireless multicarrier communications,”IEEE signal processing magazine, vol. 17, no. 3, pp. 29–48, May 2000

    work page 2000

  37. [37]

    R. G. Gallager,Information theory and reliable communication. New York, NY , USA: Wiley, 1968

    work page 1968

  38. [38]

    T. M. Cover,Elements of information theory, 1st ed. Hoboken, NJ, USA: Wiley, 1999

    work page 1999

  39. [39]

    Tse and P

    D. Tse and P. Viswanath,Fundamentals of wireless communication. Cambridge, U.K.: Cambridge university press, 2005

    work page 2005

  40. [40]

    The bussgang decomposition of nonlinear systems: Basic theory and MIMO extensions [lecture notes],

    O. T. Demir and E. Bjornson, “The bussgang decomposition of nonlinear systems: Basic theory and MIMO extensions [lecture notes],”IEEE Signal Processing Magazine, vol. 38, no. 1, pp. 131–136, Jan. 2021

    work page 2021

  41. [41]

    How much training is needed in multiple-antenna wireless links?

    B. Hassibi and B. M. Hochwald, “How much training is needed in multiple-antenna wireless links?”IEEE Transactions on Information Theory, vol. 49, no. 4, pp. 951–963, Apr. 2003

    work page 2003

  42. [42]

    Worst-case additive noise in wireless networks,

    I. Shomorony and A. S. Avestimehr, “Worst-case additive noise in wireless networks,”IEEE Transactions on Information Theory, vol. 59, no. 6, pp. 3833–3847, June 2013

    work page 2013

  43. [43]

    J. S. Liu,Monte Carlo Strategies in Scientific Computing, ser. Springer Series in Statistics. New York, NY , USA: Springer, 2001, vol. 10. 39

    work page 2001

  44. [44]

    Durrett,Probability: Theory and examples, 5th ed

    R. Durrett,Probability: Theory and examples, 5th ed. Cambridge, U.K.: Cambridge university press, 2019, vol. 49

    work page 2019

  45. [45]

    A law of large numbers for identically distributed martingale differences,

    J. Elton, “A law of large numbers for identically distributed martingale differences,”The Annals of Probability, vol. 9, no. 3, pp. 405–412, Jun. 1981

    work page 1981

  46. [46]

    R. A. Horn and C. R. Johnson,Matrix analysis. Cambridge, U.K.: Cambridge university press, 2012

    work page 2012

  47. [47]

    Cvetkovski,Inequalities: Theorems, techniques and selected problems

    Z. Cvetkovski,Inequalities: Theorems, techniques and selected problems. Berlin, Germany: Springer, 2012

    work page 2012

  48. [48]

    Doubly stochastic matrices and the diagonal of a rotation matrix,

    A. Horn, “Doubly stochastic matrices and the diagonal of a rotation matrix,”American Journal of Mathematics, vol. 76, no. 3, pp. 620–630, Jul. 1954

    work page 1954

  49. [49]

    P. J. Schreier and L. L. Scharf,Statistical signal processing of complex-valued data: The theory of improper and noncircular signals. Cambridge, UK: Cambridge University Press, 2010

    work page 2010