Efficient accessible bounds to the classical capacity of quantum channels

Chiara Macchiavello; Massimiliano F. Sacchi

REVIEW 1 major objections 5 minor 58 references

Efficient accessible bounds to the classical capacity of quantum channels

T0 review · 1 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read A few local measurements certify a lower bound on any quantum channel's classical capacity

desk verdict Sound and useful: a d^2-measurement lower bound for classical capacity, but the experimental 'accessible' claim needs a finite-sample error analysis. read the letter →

arxiv 1908.01614 v2 pith:WYYER7YD submitted 2019-08-05 quant-ph

Chiara Macchiavello , Massimiliano F. Sacchi This is my paper · ORCID

classification quant-ph PACS 03.67.Hk

keywords classicalcapacityquantumchannellowerboundlocalmeasurementsHolevoprocesstomographyBlahut-Arimotoalgorithmqubitchannels

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 presents a method to certify a lower bound on the classical capacity of a noisy quantum communication channel without performing full process tomography. The method prepares a system qudit and a noiseless reference qudit in a maximally entangled state, measures a small set of local observables of the form $X_i \otimes X_i^T$, and reconstructs the conditional probabilities of the channel acting on the eigenstates of each $X_i$. A classical optimization over input priors, using the Blahut–Arimoto recursion, yields a mutual information $I^{(i)}$, and the detected bound is $C_{\rm DET} = \max_i I^{(i)}$, which provably satisfies $C(\mathcal{E}) \ge C_1(\mathcal{E}) \ge C_{\rm DET}$. The number of measurement settings scales at most as $d^2$ for a $d$-dimensional system, requires no a priori knowledge of the channel, and the protocol returns an explicit encoding that attains the detected bound. For all unital qubit channels and for nonunital qubit channels satisfying the pseudoclassicality condition, this detected bound is exactly the Holevo capacity.

What carries the argument

The load-bearing identity is $\mathrm{Tr}[(A \otimes B^T)(\mathcal{E} \otimes I_R)|\varphi^+\rangle\langle\varphi^+|] = \frac{1}{d}\mathrm{Tr}[A\,\mathcal{E}(B)]$, which converts local measurements of $X_i \otimes X_i^T$ on the entangled state into the channel's transition probabilities. The classical optimization of the mutual information $I^{(i)}$ over the input prior is performed with the Blahut–Arimoto recursion, guaranteeing convergence to the optimal prior for each measured basis. The detection bound is then $C_{\rm DET} = \max_i I^{(i)}$, and the paper shows this quantity satisfies $C \ge C_1 \ge C_{\rm DET}$, where $C_1$ is the Holevo capacity.

What would settle it

Run the protocol on a channel whose Holevo capacity is known analytically but whose optimal ensemble is nonorthogonal (e.g., a qubit channel violating pseudoclassicality), using finite measurement statistics, and check whether the estimated $C_{\rm DET}$ ever exceeds the known $C_1$; exceeding it would invalidate the bound as an experimental guarantee. Alternatively, add a small depolarizing noise to the reference system and observe whether the reconstructed probabilities break the assumptions of the trace identity.

Watch

Extended reading notes

Core claim

For any completely positive trace-preserving quantum channel $\mathcal{E}$ on a $d$-dimensional system, the authors construct a quantity $C_{\rm DET} = \max_i I^{(i)}$ from conditional probabilities $p^{(i)}(m|n) = \langle \varphi^{(i)}_m | \mathcal{E}(|\varphi^{(i)}_n \rangle \langle \varphi^{(i)}_n|) | \varphi^{(i)}_m \rangle$, which are obtained by measuring local observables $X_i \otimes X_i^T$ on a maximally entangled state and using the identity (3). They prove the chain $C(\mathcal{E}) \ge C_1(\mathcal{E}) \ge C_{\rm DET}$, establishing $C_{\rm DET}$ as an experimentally accessible lower bound to the classical capacity. The bound improves monotonically as more observables are measured, and for important qubit-channel classes — all unital qubit channels and pseudoclassical nonunital channels — it saturates to the Holevo capacity, so the certification is tight in those cases.

Load-bearing premise

The detection scheme presupposes a noiseless reference system and ideal projective measurements; under real finite statistics the reconstructed conditional probabilities are noisy, and the paper does not analyze how statistical overestimates could make the detected bound exceed the true capacity.

Editorial extensions

If this is right

For a completely unknown qubit or qutrit channel, the protocol certifies a lower bound on classical capacity using only 1 to $d^2-1$ local measurement settings, whereas full process tomography requires $d^4$ parameters.
For all unital qubit channels the detected bound is tight: $C_{\rm DET} = C_1 = C$.
For nonunital qubit channels satisfying the pseudoclassicality condition — the shift vector parallel to a principal axis and a sufficiently thin ellipsoid — the detected bound again equals the Holevo capacity.
When the measurement bases are badly matched to the unknown channel, the bound can be loose; for a dephasing channel on an unknown basis the worst-case detected capacity is $1-H(2p/3)$, illustrating the trade-off between basis choice and tightness.
The protocol outputs an explicit encoding (input states and prior probabilities) that achieves the detected lower bound, so it can be directly used to implement a communication scheme.

Reading between the lines

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

Because $C_{\rm DET}$ is a maximum over measured bases and optimized priors, finite measurement statistics can in principle push the estimated value above the true capacity; a rigorous experimental protocol would need to report confidence intervals or use a penalized estimator.
The same local correlation measurements could be reused to certify other capacity-like quantities, such as a lower bound on the private classical capacity via the known inequality chain, though this may be considerably looser.
An adaptive strategy that chooses the next observable $X_i$ based on previously estimated transition probabilities could approach the Holevo capacity with fewer measurement settings than a fixed set of bases.
Applying the method to random measurement bases for high-dimensional channels could give a statistical estimate of how many bases are needed to get close to the Holevo capacity, since the bound is monotonically nondecreasing with added bases.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Colleague,

The paper is a neat, honest extension of the authors' QDET detection method to classical capacity. The core chain C(E) ≥ C1(E) ≥ C_DET = max_i I^(i) is correct: reconstructing the conditional probabilities via the maximally entangled state identity (3) and then computing mutual information for each basis gives a genuine lower bound, since each I^(i) corresponds to a specific encoding-decoding scheme. The math is standard and the examples check out. For unital qubit channels, C_DET equals the Holevo capacity because they are pseudoclassical; for nonunital channels the paper gives analytic formulas and honestly notes when the bound is loose. The qutrit V-shaped channel example is a good demonstration that the method can produce nontrivial bounds for channels where the capacity isn't well known.

What's genuinely new is the adaptation to classical capacity with d^2 measurement scaling, plus the explicit reconstruction recipe and the new analytic expressions. The citation pattern is appropriate, including the self-citations, since this builds directly on their QDET work. No circularity.

The soft spot is finite-sample statistics. C_DET is a maximum over bases and priors, and mutual information is convex in the transition matrix. If you plug in an unbiased estimate of p^(i)(m|n) from finite counts, Jensen's inequality gives an upward bias in the estimated C_DET; for a completely depolarizing channel, noise alone will produce a positive number with high probability. The paper never mentions this. The ideal-probability theorem is fine, but the claim that the method 'detects' a certified lower bound in practice is not established without confidence intervals or sample-complexity bounds. This doesn't sink the paper, but it should be stated as a limitation or addressed in a follow-up. I'd also note that the bound tests only orthogonal ensembles, so there are channels where it's far below C1; the paper acknowledges this.

Overall, this is a solid Letter for experimentalists benchmarking unknown channels and for theorists interested in detection methods. It deserves a serious referee; I'd ask for a finite-size error analysis or an explicit caveat in the main text. Worth a quick read at the next group meeting.

Referee Report

1 major / 5 minor

Summary. The paper proposes a method to obtain a lower bound on the classical capacity of an unknown quantum channel without full process tomography. The scheme prepares a maximally entangled state, applies the channel to the system half, measures local observables of the form X_i ⊗ X_i^T, and reconstructs conditional probabilities p^(i)(m|n) using the identity in Eq. (3). From these probabilities one computes I^(i), the capacity of the induced classical binary/dit channel, and defines C_DET as the maximum over i. The chain C ≥ C1 ≥ C_DET in Eq. (5) is the central result. The paper then derives explicit formulas for Pauli channels, generalized amplitude-damping channels, stretched-damping channels, extremal qubit channels, a qutrit V-shaped decay channel, dephasing on an unknown basis, and Pauli channels followed by an unknown phase rotation, using both analytic and numerical (Blahut–Arimoto) optimization.

Significance. The mathematical core is a clean and correct inequality: for exact conditional probabilities, the mutual information achievable with any fixed orthonormal input ensemble and output measurement is a lower bound on the Holevo capacity, and hence on the classical capacity. The identity in Eq. (3) provides a practical way to extract these conditional probabilities with local measurements. If the paper's protocol is implemented with rigorous statistical treatment, it would offer a valuable experimental tool for certifying a guaranteed lower bound on classical capacity for completely unknown channels, complementing the authors' earlier QDET bound for quantum capacity. The examples are well chosen and demonstrate that the bound is often tight, sometimes exactly equal to the true capacity.

major comments (1)

[Detection strategy, Eq. (5)] The derivation of Eq. (5) is sound for exact conditional probabilities p^(i)(m|n), but the paper's claim that the method provides an 'experimentally accessible bound' is not justified under finite statistics. In practice the probabilities are estimated from a finite number of trials, and the estimated C_DET is a maximum over i and over the input prior. Because the capacity of a classical discrete memoryless channel is a convex function of its transition matrix, Jensen's inequality gives E[C(p-hat)] ≥ C(p), so the plug-in estimate is biased upward. For a zero-capacity channel, any statistical noise produces a positive estimated bound with high probability. The paper does not provide a confidence-interval construction, a one-sided test, or a sample-complexity bound. Without such a correction, the protocol does not certify a lower bound in a real experiment. I recommend adding a paragraph (or a short section) that either gives a finite-size confidence-interval procedure or explicitly states that the bound applies to ideal probabilities and that a rigorous statistical extension is needed.

minor comments (5)

[Supplemental Material, Eq. (A.3)] The definition of z in Eq. (A.3) is ambiguous in the typeset text: it should read z = 2^{(H[ϵ0]−H[ϵ1])/(1−ϵ0−ϵ1)}, with the parentheses in the exponent made explicit.
[Eq. (12) and surrounding text] The statement that C = C1 = CDET for Pauli channels rests on additivity of the Holevo capacity for unital qubit channels, but the cited reference [46] is not the standard theorem for qubit additivity; the authors should cite King's theorem on additivity for all qubit channels (or for unital qubit channels specifically) to make the justification precise.
[Generalized amplitude-damping example, Eq. (17)] The claim 'We numerically checked that the maximum is always achieved by the first term in Eq. (17)' is a strong statement over the full parameter range; a brief analytic argument or a plot of the difference would make the example more convincing.
[Extremal qubit channels, Eq. (A.8)] Similarly, the numerical check that the first term in Eq. (A.8) is always the maximizer is reported without supporting data; a short proof or figure would strengthen the example.
[General remarks] The phrase 'number of local measurements that scales at most as d^2' should be clarified: each measurement setting X_i ⊗ X_i^T yields d^2 joint probabilities, so d^2 settings produce d^4 numbers, which is already enough for full process tomography. The advantage of the method is that a smaller number of settings still yields a nontrivial bound; this distinction should be stated explicitly to avoid a misleading scalability claim.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: C_DET is an independent lower bound derived from standard Holevo inequalities.

full rationale

The derivation of Eq. (5) is self-contained and not circular. The detection bound C_DET is defined as max_i I^(i), where I^(i) is the mutual information of the conditional probability matrix p^(i)(m|n) reconstructed from local measurements. The chain C(E) >= C1(E) >= C_DET follows from the Holevo-Schumacher-Westmoreland theorem and from the fact that any mutual information achievable with orthogonal encodings and fixed output measurements is a valid lower bound on C1; neither side of the inequality is defined in terms of the other. The measurement reconstruction uses the identity Tr[(A⊗B^T)(E⊗I_R)|φ^+⟩⟨φ^+|] = (1/d) Tr[A E(B)], which is a standard mathematical identity, not an assumption equivalent to the target result. Self-citations to the authors' earlier QDET method (Refs. [24-27]) are contextual: the paper explicitly notes QDET gives a bound on C but is 'in general very loose', and the new C_DET bound is constructed independently from conditional probabilities and classical optimization. Benchmarks against known Holevo capacities use externally established results on additivity and pseudoclassical channels. There is no fitted parameter renamed as a prediction and no self-citation invoked to forbid alternatives. The only caveat is finite-sample bias in estimating p^(i)(m|n), which is an experimental certification issue, not circularity in the mathematical derivation.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The method introduces no new physical entities and no fitted parameters. The central bound follows from standard coding theorems and the reconstruction identity. All examples use pre-specified channel parameters as inputs, not as fit parameters.

assumptions (3)

domain assumption The channel is a memoryless completely positive trace-preserving map; the classical capacity is the regularized limit of Holevo capacities and the single-letter Holevo capacity C1 is a lower bound on C.
Invoked in Eq. (5) via the chain C >= C1 >= C_DET; it is standard for quantum channel coding.
domain assumption The identity Tr[(A⊗B^T)(E⊗I_R)|φ+><φ+|] = 1/d Tr[A E(B)] in Eq. (3) holds, requiring a noiseless reference system and a fixed basis for the transpose.
This is the measurement reconstruction step; if the reference is noisy, the measured probabilities do not match p^(i)(m|n).
standard math The Blahut-Arimoto algorithm converges to the global maximum of mutual information over the input prior for a fixed transition matrix.
Used in Eq. (6) to compute I^(i) in Eq. (4).

how reviews work

0 comments

Cite this review

Pith. "Pith review of Efficient accessible bounds to the classical capacity of quantum channels." pith.science (2026). https://pith.science/paper/WYYER7YD

@misc{pith2026190801614,
  author       = {Pith},
  title        = {Pith review of: Efficient accessible bounds to the classical capacity of quantum channels},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WYYER7YD}},
  note         = {Machine review of arXiv:1908.01614}
}

read the original abstract

We present a method to detect lower bounds to the classical capacity of quantum communication channels by means of few local measurements (i.e. without complete process tomography), reconstruction of sets of conditional probabilities, and classical optimisation. The method does not require any a priori information about the channel. We illustrate its performance for significant forms of noisy channels.

Figures

Figures reproduced from arXiv: 1908.01614 by the authors.

**Figure 1.** FIG. 1. Detected classical capacity [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗

**Figure 1.** FIG. 1. Detected classical capacity [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗

Figures from the paper (3 more)

**Figure 2.** Figure 2: FIG. 2. Decay processes for a three-level atom with V-shaped pattern [PITH_FULL_IMAGE:figures/full_fig_p007_2.png]

**Figure 3.** Figure 3: FIG. 3. Detected classical capacity for a qubit dephasing channel on unknown basis with error probability [PITH_FULL_IMAGE:figures/full_fig_p009_3.png]

**Figure 4.** Figure 4: FIG. 4. Average detected classical capacity for a Pauli channel with error propabilities [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

58 extracted references · 55 canonical work pages

[1]

M. A. Nielsen and I. L. Chuang, Quantum Information and Communication (Cambridge, Cambridge University Press, 2000)

work page 2000
[2]

That is, the channel Φ produces the aﬃne mapping⃗ r↦→ Λ⃗ r+⃗t. Via local unitary operations acting before and after the map, the transformation matrices Λ and⃗t may be brought to the form Λ =   λ1 0 0 0 λ2 0 0 0 λ3  , ⃗t =   t1 t2 t3  , (9) namely, an arbitrary qubit channel Φ may be expressed as Φ = ΓU◦Φ⃗t,Λ◦ΓV , where ΓU and ΓV are unitary chann...

work page
[3]

A. S. Holevo, Prob. Inf. Transm. 9, 177 (1973)

work page 1973
[4]

Schumacher and M

B. Schumacher and M. D. Westmoreland, Phys. Rev. A 56, 131 (1997)

work page 1997
[5]

A. S. Holevo, IEEE. Trans. Inf. Theory 44, 269 (1998)

work page 1998
[6]

H. Imai, M. Hachimori, M. Hamada, H. Kobayashi and K. Matsumoto, Proceedings of the 2nd Japanese- Hungarian Symposium on Discrete Mathematics and Its Applications, p. 60 (Budapest, 2001)

work page 2001
[7]

Osawa and H

S. Osawa and H. Nagaoka, IEICE Trans. on Fundamen- tals of Electronics, Communications and Computer Sci- ences E84, 2583 (2001)

work page 2001
[8]

Hayashi, H

M. Hayashi, H. Imai, K. Matsumoto, M. B. Ruskai, and T. Shimono, Quantum Inf. Comput. 5, 13 (2005)

work page 2005

Show all 58 references

[9]

Sutter, D

T. Sutter, D. Sutter, P. M. Esfahani, and J. Lygeros, IEEE Trans. Inf. Theory 61, 1649 (2016)

2016
[10]

Sutter, T

D. Sutter, T. Sutter, P. M. Esfahani, and R. Renner, IEEE Trans. Inf. Theory 62, 578 (2016)

2016
[11]

I. L. Chuang and M. A. Nielsen, Journal of Modern Op- tics 44, 2455 (1997)

1997
[12]

J. F. Poyatos, J. I. Cirac, and P. Zoller, Phys. Rev. Lett. 78, 390 (1997)

1997
[13]

M. F. Sacchi, Phys. Rev. A 63, 054104 (2001)

2001
[14]

G. M. D’Ariano and P. Lo Presti, Phys. Rev. Lett. 86, 4195 (2001)

2001
[15]

Altepeter, D

J. Altepeter, D. Branning, E. Jeﬀrey, T. Wei, P. Kwiat, R. Thew, J. OBrien, M. Nielsen, and A. White, Phys. Rev. Lett. 90, 193601 (2003)

2003
[16]

J. L. O’Brien, G. J. Pryde, A. Gilchrist, D. F. V. James, N. K. Langford, T. C. Ralph, and A. G. White, Phys. Rev. Lett. 93, 080502 (2004)

2004
[17]

S. H. Myrskog, J. K. Fox, M. W. Mitchell, and A. M. Steinberg, Phys. Rev. A 72, 013615 (2005)

2005
[18]

Riebe, K

M. Riebe, K. Kim, P. Schindler, T. Monz, P. O. Schmidt, T. K. K¨ orber, W. H¨ ansel, H. H¨ aﬀner, C. F. Roos, and R. Blatt, Phys. Rev. Lett. 97, 220407 (2006)

2006
[19]

Mohseni, A

M. Mohseni, A. T. Rezakhani, and D. A. Lidar, Phys. Rev. A 77, 032322 (2008)

2008
[20]

Bongioanni, L

I. Bongioanni, L. Sansoni, F. Sciarrino, G. Vallone, and P. Mataloni, Phys. Rev. A 82, 042307 (2010)

2010
[21]

Y. Sagi, I. Almog, and N. Davidson, Phys. Rev. Lett. 105, 053201 (2010)

2010
[22]

Macchiavello and M

C. Macchiavello and M. Rossi, Phys. Rev. A 88, 042335 (2013)

2013
[23]

Orieux, L

A. Orieux, L. Sansoni, M. Persechino, P. Mataloni, M. Rossi, and C. Macchiavello, Phys. Rev. Lett.111, 220501 (2013)

2013
[24]

Chruscinski, C

D. Chruscinski, C. Macchiavello, and S. Maniscalco, Phys. Rev. Lett. 118, 080404 (2017)

2017
[25]

Macchiavello and M

C. Macchiavello and M. F. Sacchi, Phys. Rev. Lett. 116, 140501 (2016)

2016
[26]

Macchiavello and M

C. Macchiavello and M. F. Sacchi, Phys. Rev. A 94, 052333 (2016)

2016
[27]

Cuevas, M

A. Cuevas, M. Proietti, M. A. Ciampini, S. Duranti, P. Mataloni, M. F. Sacchi, and C. Macchiavello, Phys. Rev. Lett. 119, 100502 (2017)

2017
[28]

Macchiavello and M

C. Macchiavello and M. F. Sacchi, Phys. Rev. A 97, 012303 (2018)

2018
[29]

M. E. Shirokov, Prob. Inf. Transm. 48, 85 (2012)

2012
[30]

C. A. Fuchs, Phys. Rev. Lett. 79, 1162 (1997)

1997
[31]

A. S. Holevo, Russian Math. Surveys 53, 1295 (1998)

1998
[32]

King and M

C. King and M. B. Ruskai, J. Math. Phys. 42, 87 (2001)

2001
[33]

G. M. D’Ariano, P. Lo Presti, and M. F. Sacchi, Phys. Lett. A 272, 32 (2000)

2000
[34]

Dall’Arno, G

M. Dall’Arno, G. M. D’Ariano, and M. F. Sacchi, Phys. Rev. A 83, 062304 (2011)

2011
[35]

R. E. Blahut, IEEE Trans. on Inf. Theory 18, 460 (1972)

1972
[36]

R. G. Gallager, Information Theory and Reliable Com- munication (John Wiley & Sons, 1968)

1968
[37]

Arimoto, IEEE Trans

S. Arimoto, IEEE Trans. on Inf. Theory 18, 14 (1972)

1972
[38]

G. M. D’Ariano and M. F. Sacchi, Opt. Commun. 149, 152 (1998)

1998
[39]

T. M. Cover and J. A. Thomas, Elements of Information Theory (Wiley & Sons, New Jersey, 2006)

2006
[40]

G. M. D’Ariano, P. Perinotti, and M. F. Sacchi, J. Opt. B 6, S487 (2004)

2004
[41]

ur Rehman, Y

J. ur Rehman, Y. Jeong, J. San Kim, and H. Shin, Sci- entiﬁc Reports 8, 17457 (2018)

2018
[42]

ur Rehman, Y

J. ur Rehman, Y. Jeong, and H. Shin, Phys. Rev. A 99, 042312 (2019)

2019
[43]

King and M

C. King and M. B. Ruskai, IEEE Trans. Info. Theory 47, 192 (2001)

2001
[44]

M. B. Ruskai, S. Szarek, and E. Werner, Lin. Alg. Appl. 347, 159 (2002)

2002
[45]

Fujiwara and P

A. Fujiwara and P. Algoet, Phys. Rev. A59, 3290 (1999)

1999
[46]

Braun, O

D. Braun, O. Giraud, I. Nechita, C. Pellegrini, and M. ˇZnidaric, J. Phys. A 47, 135302 (2014)

2014
[47]

C. King, J. Math. Phys. 43, 4641 (2002)

2002
[48]

Fujiwara and H

A. Fujiwara and H. Nagaoka, IEEE Trans. Inf. Theory 44,1071 (1998)

1998
[49]

Hayashi, Quantum Information Theory: An Introduc- tion (Springer, Berlin, 2006)

M. Hayashi, Quantum Information Theory: An Introduc- tion (Springer, Berlin, 2006)

2006
[50]

C. King, M. Nathanson, and M. B. Ruskai, Phys. Rev. 1 Lett. 88, 057901 (2002)

2002
[51]

D. W. Berry, Phys. Rev. A 71, 032334 (2005)

2005
[52]

Daems, Phys

D. Daems, Phys. Rev. Lett. 102, 180503 (2009)

2009
[53]

C. A. Fuchs and C. M. Caves, Phys. Rev. Lett. 73, 3047 (1994)

1994
[54]

C. J. Myatt, B. E. King, Q. A. Turchette, C. A. Sack- ett, D. Kielpinski, W. M. Itano, C. Monroe, and D. J. Wineland, Nature 403, 269 (2000)

2000
[55]

Q. A. Turchette, C. J. Myatt, B. E. King, C. A. Sack- ett, D. Kielpinski, W. M. Itano, C. Monroe, and D. J. Wineland, Phys. Rev. A 62, 053807 (2000)

2000
[56]

Uhlmann, J

A. Uhlmann, J. Phys. A 34, 7047 (2001)

2001
[57]

Giovannetti and R

V. Giovannetti and R. Fazio, Phys. Rev. A 71, 032314 (2005)

2005
[58]

Ch¸ eci´ nska and K

A. Ch¸ eci´ nska and K. W´ odkiewicz, Phys. Rev. A 76, 052306 (2007). SUPPLEMENT AL MA TERIAL Capacity for classical binary asymmetric channels Without loss of generality we can ﬁx the labeling of zeros and ones in a classical binary channel such that 0≤ϵ0≤ 1 2, ϵ0≤ ϵ1, and ϵ0...

2007

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [1]

M. A. Nielsen and I. L. Chuang, Quantum Information and Communication (Cambridge, Cambridge University Press, 2000)

work page 2000

[2] [2]

That is, the channel Φ produces the aﬃne mapping⃗ r↦→ Λ⃗ r+⃗t. Via local unitary operations acting before and after the map, the transformation matrices Λ and⃗t may be brought to the form Λ =   λ1 0 0 0 λ2 0 0 0 λ3  , ⃗t =   t1 t2 t3  , (9) namely, an arbitrary qubit channel Φ may be expressed as Φ = ΓU◦Φ⃗t,Λ◦ΓV , where ΓU and ΓV are unitary chann...

work page

[3] [3]

A. S. Holevo, Prob. Inf. Transm. 9, 177 (1973)

work page 1973

[4] [4]

Schumacher and M

B. Schumacher and M. D. Westmoreland, Phys. Rev. A 56, 131 (1997)

work page 1997

[5] [5]

A. S. Holevo, IEEE. Trans. Inf. Theory 44, 269 (1998)

work page 1998

[6] [6]

H. Imai, M. Hachimori, M. Hamada, H. Kobayashi and K. Matsumoto, Proceedings of the 2nd Japanese- Hungarian Symposium on Discrete Mathematics and Its Applications, p. 60 (Budapest, 2001)

work page 2001

[7] [7]

Osawa and H

S. Osawa and H. Nagaoka, IEICE Trans. on Fundamen- tals of Electronics, Communications and Computer Sci- ences E84, 2583 (2001)

work page 2001

[8] [8]

Hayashi, H

M. Hayashi, H. Imai, K. Matsumoto, M. B. Ruskai, and T. Shimono, Quantum Inf. Comput. 5, 13 (2005)

work page 2005