Pith. sign in

REVIEW 3 major objections 4 minor 42 references

Practical Short-Length Coding Schemes for Binary Distributed Hypothesis Testing

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Practical short-length coding schemes for binary distributed hypothesis testing can be built from short binary linear block codes, and the paper derives exact analytical error probabilities for them.

desk verdict The practical short-length DHT schemes and simulations are credible, but the claimed exact error-probability formulas rest on a binomial count that conditions on the wrong weight and fail direct enumeration. read the letter →

arxiv 2506.01747 v1 pith:4YGXE2WI submitted 2025-06-02 cs.IT eess.SPmath.IT

Ismaila Salihou Adamou , Elsa Dupraz , Reza Asvadi , Tad Matsumoto This is my paper
classification cs.ITeess.SPmath.IT
keywords distributedhypothesistestingshort-lengthcodesbinaryquantizationquantize-binninglinearblockNeyman-PearsonType-IerrorType-II
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

This paper sets out to make distributed hypothesis testing (DHT) practical at short block lengths. Where earlier DHT work proved asymptotic error-exponent bounds using quantizers and random binning, this paper implements those same two ingredients—quantization and quantize-binning—with short binary linear block codes (BCH and Reed-Muller codes) decoded by exact syndrome-based maximum-likelihood decoders. The central theoretical contribution is a pair of exact analytical formulas, Proposition 1 for the quantizer-alone scheme and Proposition 2 for the quantize-binning scheme, that give the Type-I and Type-II error probabilities in the asymmetric setup (Y available losslessly at the decoder) when p0 = p1 = 1/2. These formulas are expressed through the weight distribution of coset leaders of the code, and the paper reports that they match Monte-Carlo simulations of the proposed schemes. If correct, the formulas let designers evaluate and optimize DHT decision performance without running simulations, for the block lengths below 100 bits that matter in practice.

What carries the argument

The load-bearing object is the coset-leader weight distribution of the quantizer code: for the all-zero codeword, E_gamma^((q)) counts how many source words of Hamming weight gamma are quantized to the zero codeword, and N_0^((q)) normalizes these counts. The error formulas sum, over distance class j and side-information weight $\lambda$, the conditional counting term Delta_{$\lambda$,j,gamma} defined in Eq. (30) through the binomial product Gamma_{$\lambda$,j,gamma} of Eq. (31); for quantize-binning, the analogous counts E_nu^((qb)) and the concatenated-code weight distribution A_t^((qb)) play the same role. These combinatorial profiles convert the hypothesis test over a structured code into sums over weight classes, which is what makes a closed-form—and supposedly exact—evaluation possible.

What would settle it

Enumerate exhaustively, for a short code such as the length-3 repetition code or the (31,16) BCH code with p0 = p1 = 1/2, all pairs (x^n, y^n), compute by brute force the conditional probability P0(w(Y^n) = $\lambda$ | w(X^n) = gamma, d(X^n, Y^n) = j), and compare it with Delta_{$\lambda$,j,gamma} from Eqs. (30)-(31); any mismatch between the enumerated value and the formula settles whether the claimed exact expressions are exact.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the information-theoretic quantizer and quantize-binning constructions for distributed hypothesis testing have finite-length incarnations with closed-form error probabilities. For the asymmetric setup and equiprobable X under both hypotheses, the Type-I error alpha_n and Type-II error beta_n of the quantizer-alone scheme are claimed to be given exactly by Proposition 1, which depends only on n, the noise parameters c0 and c1, the threshold lambda_q, and the coset-leader weight profile E_gamma^((q)) of the quantizer code; the quantize-binning scheme is likewise claimed to be exactly described by Proposition 2 in terms of the quantizer's coset-leader counts E_gamma^((q)), the binning decision-region counts E_nu^((qb)), and the weight distribution A_t^((qb)) of the concatenated code. The same schemes, evaluated by Monte-Carlo simulation, are shown to outperform both the uncoded truncation baseline and existing LDGM-based quantizers using belief-propagation decoders at lengths n=31 and n=63. The analytical expressions are presented as new tools for optimizing and comparing practical DHT schemes without simulation.

Load-bearing premise

The exact error formulas stand or fall on the combinatorial identity in Equations (30)-(31), which counts how many side-information strings of a given weight sit at a given Hamming distance from a quantized codeword of a given weight; if that count is wrong, the printed formulas do not reproduce the true error probabilities.

Editorial extensions

If this is right

  • Code design for DHT can be turned into an optimization over the parameters E_gamma^((q)), E_nu^((qb)), N_0^((q)), and A_t^((qb)), without repeated Monte-Carlo runs.
  • At equal transmission rate, the proposed quantize-binning scheme is claimed to dominate the truncation scheme in both asymmetric and symmetric setups.
  • Short BCH and Reed-Muller codes with syndrome-based ML decoding beat LDGM codes with bias-propagation decoders at block lengths 31 and 63.
  • The analytical expressions cover the important testing-against-independence subcase c1 = 1/2 and give finite-length ROC curves for it.
  • The framework brings the SHA-style quantize-binning scheme from an asymptotic exponent bound to a directly implementable finite-length code.

Reading between the lines

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

  • A natural extension is to re-derive Propositions 1 and 2 with the roles of the quantized weight and the source weight exchanged in the conditional counting term; the resulting formulas would still be checkable by brute-force enumeration on short codes.
  • The same coset-leader machinery could yield finite-length error expressions for p0 != p1 by conditioning on the true source distribution instead of relying on the all-zero-codeword symmetry.
  • For the symmetric setup, a symmetric analogue of Proposition 2 would require a joint weight distribution over two decoded codewords, which the present paper leaves open.
  • Because the analytical formulas hinge on exact weight counts, code search could be guided by covering-radius and weight-enumerator data already tabulated for classical block codes, making the optimization step immediate.
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

3 major / 4 minor

Summary. The paper proposes practical short-length coding schemes for binary distributed hypothesis testing, using short binary linear block codes for quantization and quantize-binning in both asymmetric and symmetric setups. The authors derive what they claim are exact analytical expressions for the Type-I and Type-II error probabilities of the quantizer-alone and quantize-binning schemes for the asymmetric setup with p0 = p1 = 1/2, and validate these expressions by Monte Carlo simulations. The paper also compares the proposed schemes against uncoded separate and truncation baselines and against existing LDGM/BiP quantizers.

Significance. If the claimed exact error-probability formulas were correct, they would provide a useful tool for optimizing short-length DHT codes without Monte Carlo simulation, and the practical syndrome-based decoding implementation would be a genuine contribution to a relatively unexplored area. The empirical comparisons of short BCH and Reed-Muller codes for DHT are also of interest. However, the central theoretical claim is invalid: the combinatorial probability in Eqs. (30)-(31) is mis-specified, and Proposition 2 relies on an additional unjustified uniformity assumption. The claimed exactness therefore fails, and the reported theoretical validation cannot stand. The practical schemes and simulations may still have value, but the main advertised contribution is not sound as written.

major comments (3)
  1. [§IV-D, Eqs. (30)-(31)] The quantity Δ_{λ,j,γ} is not the conditional probability P0(w(Y^n)=λ | w(X^n)=γ, d(X^n,Y^n)=j). For a fixed X^n of weight γ, the number of Y^n of weight λ at Hamming distance j is C(γ, j-u)C(n-γ, u) with u=(j+λ-γ)/2, so the correct conditional probability is C(γ, j-u)C(n-γ, u)/C(n, j). Equation (31) instead uses C(λ,u)C(n-λ, j-u), which is the count for the reverse conditioning on w(Y^n)=λ, and Vandermonde's identity shows the denominator in Eq. (30) is just C(n,j). For n=3, γ=1, λ=3, j=2, the correct value is 1/3 while the printed ratio is 1. Direct enumeration for a length-3 repetition code in the paper's setting gives a Type-I error of 0.757, whereas the printed formula yields 0.6355. Consequently, the exact expressions in Eqs. (28)-(29) of Proposition 1 do not follow from the derivation.
  2. [§V-D, Eq. (48) and Eq. (51)] Proposition 2 inherits the same combinatorial error through Eq. (48), which uses the invalid Γ_{ν,j,γ} from Proposition 1, and therefore Eqs. (41)-(44) are not exact. In addition, Eq. (51) assumes that P0(w(X̂^n)=t, d(X̂^n,Y^n)=ν | w(Y^n)=i) factors as E(qb)_ν/C(n,ν) · A(qb)_t Γ_{i,ν,t}/C(n,i). This is equivalent to assuming that, conditional on a wrong decoding event, the incorrect codeword X̂^n is uniformly distributed over weight classes and is independent of the weight of Y^n. This assumption is not implied by linearity of the code and is false for structured linear codes in general. The claimed exactness of Proposition 2 therefore rests on an unproved and generally invalid uniformity assumption.
  3. [§VI.B and §VI.D, Figures 3 and 6] The Monte Carlo validation of the theoretical expressions cannot support the exactness claim once Eqs. (30)-(31) are wrong. If the printed formulas happen to match simulations for the specific BCH and Reed-Muller parameters shown, that match cannot be attributed to the derived expressions; the paper provides no argument that the erroneous formula is an approximation that becomes accurate for these parameters. Since exact analytical error-probability expressions are the central advertised contribution, this is a load-bearing failure of the manuscript rather than a presentation issue.
minor comments (4)
  1. [§V-B, Eq. (37)] The second log term in Eq. (37) appears to contain a typographical error: the argument should presumably be [ĉ1,b(1-ĉ0,b)]/[ĉ0,b(1-ĉ1,b)] to mirror the form of Eq. (26), rather than using ĉ0,b in the numerator.
  2. [§III-B, Eq. (18)] Threshold notation is inconsistent: the text refers to a relationship between μ and τl, but the threshold in Eq. (18) is denoted τt.
  3. [§IV-B] The sentence 'Althoug xnq and vnq are not strictly i.i.d. in practice' contains a typo ('Althoug' for 'Although').
  4. [§V-D, Proposition 2] The parameter δ is used as the argument of PB(δ) and P\bar{B}(δ), but δ was also used as a binning parameter in Eq. (9); this overloaded notation may confuse readers.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the exact error-probability formulas are derived from code parameters and source model, not fitted to their own predictions.

full rationale

The paper's central exactness claims (Propositions 1 and 2) are derived from the source model, the code's coset-leader weights Eγ and Eν^qb, and combinatorial counting, not from fitted outputs. The Monte-Carlo-estimated parameters p̂ and ĉ are used only to instantiate the practical NP tests in (26) and (37); they are not substituted into the analytical Type-I/Type-II expressions (28)-(29) and (41)-(44), which use the true c0, c1 and the enumerated code parameters. The numerical validation in Figures 3 and 6 compares analytically evaluated curves against independent Monte-Carlo simulations, which is a legitimate check rather than a circular fit. The paper's self-citations ([14] on non-i.i.d. sources and [42] on covering codes) are contextual or illustrative and do not carry the derivation. The possible combinatorial miscount in Eq. (31) and the weight-uniformity assumption in Eq. (51) are correctness concerns, not circularity: they do not assume the target error probabilities as inputs. Therefore, no circular step is exhibited.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central derivations rest on the binary additive-noise source model, the NP lemma, and two modeling assumptions that are not exactly true: the i.i.d. Bernoulli approximation of quantized vectors and the uniform distribution of the Voronoi region across weight classes. The thresholds and Monte-Carlo-estimated test parameters are free design inputs. No new physical or mathematical entities are introduced.

free parameters (4)
  • λq = integer threshold for quantization scheme test
    Chosen to satisfy the Type-I error constraint; the analytical error expressions are parameterized by this threshold.
  • λqb = integer threshold for quantize-binning test
    Same role for the quantize-binning scheme.
  • p-hat0, c-hat0, p-hat1, c-hat1 = Monte-Carlo estimates
    Used in the Neyman-Pearson test (26) and (37); estimated by simulation, so the test is calibrated to the source model through simulation.
  • Code selection (e.g., BCH(31,16), RM(16,5)) = BCH and Reed-Muller parameters
    The specific short linear block codes are chosen by hand; the paper notes code optimization is left for future work, so the reported gains depend on these choices.
assumptions (5)
  • domain assumption The sources X and Y are i.i.d. Bernoulli sequences with Y = X ⊕ Z, Z independent of X, with p0 ≤ p1 and c0 ≤ c1.
    Section II-B, Eq. (2). This binary-additive-noise model underlies all derivations.
  • standard math The Neyman-Pearson lemma gives the optimal threshold test for the stated error constraints.
    Sections III-B, IV-B; standard result, cited as [38].
  • ad hoc to paper The quantized vectors Xq and Vq = Xq ⊕ Y are modeled as i.i.d. Bernoulli with parameters estimated by Monte Carlo.
    Section IV-B: 'we model the vectors ... as realizations of i.i.d. Bernoulli random variables'; the paper admits this is an approximation and justifies it only via numerical results.
  • domain assumption The all-zero codeword analysis is sufficient because of symmetry of the linear code and the test.
    Section IV-D: 'Due to symmetry, the quantization error probability is independent of the transmitted codeword.'
  • ad hoc to paper The Voronoi region of the all-zero codeword is uniformly distributed over weight classes in the quantize-binning analysis.
    Section V-D, Eqs. (49)-(51): the product count E_ν^(qb)/C(n,ν) * A_t^(qb) Γ_{i,ν,t}/C(n,i) treats the decision region as a random subset of each weight class; this is not exact for structured linear codes.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Practical Short-Length Coding Schemes for Binary Distributed Hypothesis Testing." pith.science (2026). https://pith.science/paper/4YGXE2WI

@misc{pith2026250601747,
  author       = {Pith},
  title        = {Pith review of: Practical Short-Length Coding Schemes for Binary Distributed Hypothesis Testing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4YGXE2WI}},
  note         = {Machine review of arXiv:2506.01747}
}
read the original abstract

This paper addresses the design of practical shortlength coding schemes for Distributed Hypothesis Testing (DHT). While most prior work on DHT has focused on informationtheoretic analyses, deriving bounds on Type-II error exponents via achievability schemes based on quantization and quantizebinning, the practical implementation of DHT coding schemes has remained largely unexplored. Moreover, existing practical coding solutions for quantization and quantize-binning approaches were developed for source reconstruction tasks considering very long code length, and they are not directly applicable to DHT. In this context, this paper introduces efficient shortlength implementations of quantization and quantize-binning schemes for DHT, constructed from short binary linear block codes. Numerical results show the efficiency of the proposed coding schemes compared to uncoded cases and to existing schemes initially developed for data reconstruction. In addition to practical code design, the paper derives exact analytical expressions for the Type-I and Type-II error probabilities associated with each proposed scheme. The provided analytical expressions are shown to predict accurately the practical performance measured from Monte-Carlo simulations of the proposed schemes. These theoretical results are novel and offer a useful framework for optimizing and comparing practical DHT schemes across a wide range of source and code parameters.

Figures

Figures reproduced from arXiv: 2506.01747 by the authors.

Figure 1
Figure 1. Distributed hypothesis testing scheme X. Let p = P(X = 1) and c = P(Z = 1), where 0 < c ≤ 1/2. The hypotheses provided in (1) can then be expressed as: H0 :(p = p0, c = c0) H1 :(p = p1, c = c1). (2) For convenience, we assume that 0 < p0 ≤ 0.5, p0 ≤ p1, and c0 ≤ c1. This model has been studied from an Information￾Theoretic perspective in [23], [36]. Notably, when p0 = p1 and c1 = 1/2, the problem reduces to testing … view at source ↗
Figure 2
Figure 2. ROC curves for separate and truncation schemes. [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. ROC curve for the (31, 16) BCH code used as a quan￾tizer, compared to the truncation scheme in the asymmetric setup [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: ROC curve for the (31, 16) BCH code used as a quantizer, compared to the truncation scheme in the symmet￾ric setup. For the practical quantizer, we consider both our syndrome-based decoder and the BiP decoder from [32] The performance of the quantization scheme with th…
Figure 6
Figure 6. Figure 6: ROC curve for the quantize-binning scheme built from [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]
Figure 7
Figure 7. Figure 7: ROC curve for the quantize-binning scheme built from [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

42 extracted references · 41 canonical work pages

  1. [1]

    The rate-distortion function for source coding with side information at the decoder,

    A. Wyner and J. Ziv, “The rate-distortion function for source coding with side information at the decoder,” IEEE Transactions on Information Theory, vol. 22, no. 1, pp. 1–10, 1976

    work page 1976

  2. [2]

    Quantization,

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

    work page 1998

  3. [3]

    6g networks: Beyond shannon towards semantic and goal-oriented communications,

    E. C. Strinati and S. Barbarossa, “6g networks: Beyond shannon towards semantic and goal-oriented communications,” Computer Networks , vol. 190, p. 107930, 2021

    work page 2021

  4. [4]

    A rate distortion approach to goal-oriented communication,

    P. A. Stavrou and M. Kountouris, “A rate distortion approach to goal-oriented communication,” in IEEE International Symposium on Information Theory (ISIT) , 2022, pp. 590–595

    work page 2022

  5. [5]

    Goal- oriented quantization: Analysis, design, and application to resource allocation,

    H. Zou, C. Zhang, S. Lasaulce, L. Saludjian, and H. V . Poor, “Goal- oriented quantization: Analysis, design, and application to resource allocation,” IEEE Journal on Selected Areas in Communications , vol. 41, no. 1, pp. 42–54, 2022

    work page 2022

  6. [6]

    Embedded sensor systems for health–a step towards personalized health,

    M. Lind ´en and M. Bj ¨orkman, “Embedded sensor systems for health–a step towards personalized health,” in pHealth 2018 . IOS Press, 2018, pp. 69–74

    work page 2018

  7. [7]

    On the detection capabilities of underwater distributed acoustic sensing,

    I. Lior, A. Sladen, D. Rivet, J.-P. Ampuero, Y . Hello, C. Becerril, H. F. Martins, P. Lamare, C. Jestin, S. Tsagkli et al. , “On the detection capabilities of underwater distributed acoustic sensing,” Journal of Geophysical Research: Solid Earth , vol. 126, no. 3, p. e2020JB020925, 2021

    work page 2021

  8. [8]

    Distributed and collaborative system to improve traf- fic conditions using fuzzy logic and v2x communications,

    J. A. S ´anchez, D. Melendi, R. Garc ´ıa, X. G. Pa ˜neda, V . Corcoba, and D. Garc ´ıa, “Distributed and collaborative system to improve traf- fic conditions using fuzzy logic and v2x communications,” V ehicular Communications, vol. 47, p. 100746, 2024

    work page 2024

Show all 42 references
  1. [9]

    Hypothesis testing with communication constraints,

    R. Ahlswede and I. Csisz ´ar, “Hypothesis testing with communication constraints,” IEEE Transactions on Information Theory , vol. 32, no. 4, pp. 533–542, 1986

    1986

  2. [10]

    Hypothesis Testing with Multiterminal Data Compression,

    T. S. Han, “Hypothesis Testing with Multiterminal Data Compression,” IEEE Transactions on Information Theory , vol. 33, no. 6, pp. 759–772, 1987

    1987

  3. [11]

    Statistical inference under multiterminal data compression,

    T. S. Han and S. Amari, “Statistical inference under multiterminal data compression,” IEEE Transactions on Information Theory , vol. 44, no. 6, pp. 2300–2324, 1998

    1998

  4. [12]

    Error bound of hypothesis testing with data compression,

    H. Shimokawa, T. S. Han, and S. Amari, “Error bound of hypothesis testing with data compression,” in IEEE International Symposium on Information Theory (ISIT) , 1994, p. 114

    1994

  5. [13]

    On the neces- sity of binning for the distributed hypothesis testing problem,

    G. Katz, P. Piantanida, R. Couillet, and M. Debbah, “On the neces- sity of binning for the distributed hypothesis testing problem,” IEEE International Symposium on Information Theory (ISIT) , pp. 2797–2801, 2015

    2015

  6. [14]

    An information- spectrum approach to distributed hypothesis testing for general sources,

    I. Salihou Adamou, E. Dupraz, and T. Matsumoto, “An information- spectrum approach to distributed hypothesis testing for general sources,” in International Zurich Seminar on Information and Communication (IZS). Proceedings. ETH Z ¨urich, 2024, pp. 144–148

    2024

  7. [15]

    Improved random-binning exponent for distributed hypothesis testing,

    Y . Kochman and L. Wang, “Improved random-binning exponent for distributed hypothesis testing,” arXiv preprint arXiv:2306.14499 , 2023

    2023 arXiv

  8. [16]

    Distributed hypothesis testing over dis- crete memoryless channels,

    S. Sreekumar and D. G ¨und¨uz, “Distributed hypothesis testing over dis- crete memoryless channels,” IEEE Transactions on Information Theory , vol. 66, no. 4, pp. 2044–2066, 2019

    2019

  9. [17]

    Distributed hypothesis testing over multi-access channels,

    S. Salehkalaibar and M. Wigger, “Distributed hypothesis testing over multi-access channels,” in IEEE Global Communications Conference (Globecom), 2018, pp. 1–6

    2018

  10. [18]

    Hypothesis testing over the two-hop relay network,

    S. Salehkalaibar, M. Wigger, and L. Wang, “Hypothesis testing over the two-hop relay network,” IEEE Transactions on Information Theory , vol. 65, no. 7, pp. 4411–4433, 2019

    2019

  11. [19]

    Multiterminal detection with zero- rate data compression,

    H. M. Shalaby and A. Papamarcou, “Multiterminal detection with zero- rate data compression,” IEEE Transactions on Information Theory , vol. 38, no. 2, pp. 254–267, 1992

    1992

  12. [20]

    Exponential-type error probabilities for multiterminal hypothesis testing,

    T. S. Han and K. Kobayashi, “Exponential-type error probabilities for multiterminal hypothesis testing,” IEEE Transactions on Information Theory, vol. 35, no. 1, pp. 2–14, 2006

    2006

  13. [21]

    Statistical inference under multiterminal rate restrictions: A differential geometric approach,

    S.-I. Amari and T. S. Han, “Statistical inference under multiterminal rate restrictions: A differential geometric approach,” IEEE Transactions on Information Theory , vol. 35, no. 2, pp. 217–227, 1989. 12

    1989

  14. [22]

    Neyman–pearson test for zero-rate multiterminal hypoth- esis testing,

    S. Watanabe, “Neyman–pearson test for zero-rate multiterminal hypoth- esis testing,” IEEE Transactions on Information Theory , vol. 64, no. 7, pp. 4923–4939, 2017

    2017

  15. [23]

    On Binary Distributed Hypothesis Testing,

    E. Haim and Y . Kochman, “On Binary Distributed Hypothesis Testing,” pp. 1–37, 2017. [Online]. Available: http://arxiv.org/abs/1801.00310

    2017 arXiv

  16. [24]

    Binary quantization using belief propagation with decimation over factor graphs of LDGM codes,

    J. Fridrich and T. Filler, “Binary quantization using belief propagation with decimation over factor graphs of LDGM codes,” in Proceedings of the 45th Allerton Conference on Coding, Communication, and Control , 2007, pp. 495–501

    2007

  17. [25]

    Distributed source coding for sensor networks,

    Z. Xiong, A. D. Liveris, and S. Cheng, “Distributed source coding for sensor networks,” IEEE Signal Processing Magazine , vol. 21, no. 5, pp. 80–94, 2004

    2004

  18. [26]

    Compression of binary sources with side information at the decoder using LDPC codes,

    A. D. Liveris, Z. Xiong, and C. N. Georghiades, “Compression of binary sources with side information at the decoder using LDPC codes,” IEEE Communications Letters , vol. 6, no. 10, pp. 440–442, 2002

    2002

  19. [27]

    Improved decoding for binary source coding with coded side information,

    A. Savard and C. Weidmann, “Improved decoding for binary source coding with coded side information,” in IEEE Information Theory Workshop (ITW), 2013, pp. 1–5

    2013

  20. [28]

    Optimized rate-adaptive protograph-based LDPC codes for source coding with side information,

    F. Ye, E. Dupraz, Z. Mheich, and K. Amis, “Optimized rate-adaptive protograph-based LDPC codes for source coding with side information,” IEEE Transactions on Communications , vol. 67, no. 6, pp. 3879–3889, 2019

    2019

  21. [29]

    Low-density graph codes that are optimal for binning and coding with side information,

    M. J. Wainwright and E. Martinian, “Low-density graph codes that are optimal for binning and coding with side information,” IEEE Transactions on Information Theory , vol. 55, no. 3, pp. 1061–1079, 2009

    2009

  22. [30]

    A coding theorem for lossy data com- pression by LDPC codes,

    Y . Matsunaga and H. Yamamoto, “A coding theorem for lossy data com- pression by LDPC codes,” IEEE Transactions on Information Theory , vol. 49, no. 9, pp. 2225–2229, 2003

    2003

  23. [31]

    Lossy source compres- sion using low-density generator matrix codes: Analysis and algorithms,

    M. J. Wainwright, E. Maneva, and E. Martinian, “Lossy source compres- sion using low-density generator matrix codes: Analysis and algorithms,” IEEE Transactions on Information Theory , vol. 56, no. 3, pp. 1351– 1368, 2010

    2010

  24. [32]

    Lossy source coding using belief prop- agation and soft-decimation over LDGM codes,

    D. Castanheira and A. Gameiro, “Lossy source coding using belief prop- agation and soft-decimation over LDGM codes,” in IEEE International Symposium on Personal, Indoor and Mobile Radio Communications , 2010, pp. 431–436

    2010

  25. [33]

    Binary Wyner–Ziv code design based on compound LDGM–LDPC structures,

    M. Nangir, M. Ahmadian-Attari, and R. Asvadi, “Binary Wyner–Ziv code design based on compound LDGM–LDPC structures,” IET Com- munications, vol. 12, no. 4, pp. 375–383, 2018

    2018

  26. [34]

    Transmission of noisy information to a noisy receiver with minimum distortion,

    J. Wolf and J. Ziv, “Transmission of noisy information to a noisy receiver with minimum distortion,” IEEE Transactions on Information Theory , vol. 16, no. 4, pp. 406–411, 1970

    1970

  27. [35]

    A rate-distortion framework for characterizing semantic information,

    J. Liu, W. Zhang, and H. V . Poor, “A rate-distortion framework for characterizing semantic information,” in IEEE International Symposium on Information Theory (ISIT) , 2021, pp. 2894–2899

    2021

  28. [36]

    Distributed Binary Detection with Lossy Data Compression,

    G. Katz, P. Piantanida, and M. Debbah, “Distributed Binary Detection with Lossy Data Compression,” IEEE Transactions on Information Theory, vol. 63, no. 8, pp. 5207–5227, 2017

    2017

  29. [37]

    On the optimality of binning for distributed hypothesis testing,

    M. S. Rahman and A. B. Wagner, “On the optimality of binning for distributed hypothesis testing,” IEEE Transactions on Information Theory, vol. 58, no. 10, pp. 6282–6303, 2012

    2012

  30. [38]

    E. L. Lehmann, J. P. Romano, and G. Casella, Testing statistical hypotheses. Springer, 2005, vol. 3

    2005

  31. [39]

    Information embedding codes on graphs with iterative encoding and decoding,

    V . Chandar, E. Martinian, and G. W. Wornell, “Information embedding codes on graphs with iterative encoding and decoding,” in IEEE Inter- national Symposium on Information Theory , 2006, pp. 866–870

    2006

  32. [40]

    Richardson and R

    T. Richardson and R. Urbanke, Modern Coding Theory . Cambridge university press, 2008

    2008

  33. [41]

    Ryan and S

    W. Ryan and S. Lin, Channel Codes: Classical and Modern . Cambridge university press, 2009

    2009

  34. [42]

    Covering codes as near-optimal quantizers for distributed hypothesis testing against independence,

    F. Khaledian, R. Asvadi, E. Dupraz, and T. Matsumoto, “Covering codes as near-optimal quantizers for distributed hypothesis testing against independence,” in IEEE Information Theory Workshop (ITW) , 2024, pp. 67–72

    2024

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.