arXiv:2610.03706
Quantum estimation, channel orders, and private capacity
Christoph Hirche
Abstract
Recent examples have shown that zero private capacity need not imply antidegradability and that two channels with zero private capacity can nevertheless transmit private information together. Existing entropic channel orders compare what a receiver and its environment can learn and thereby bound capacities. We connect these orders to the recent constructions through binary estimation, whose minimum mean-square error is governed by measured $χ^2$ divergence. A new integral representation shows how measured $χ^2$ on a qubit extension recovers quantum $χ^2$. It lets us pass from complete measured-$χ^2$ ordering to complete quantum-$χ^2$, relative-entropy, and less-noisy ordering. Zhu and Wang used a signed lift to show that their qutrit channel has zero private capacity. We show that Hermitian-smoothed measured-$χ^2$ ordering characterizes when such lifts exist, even with a quantum reference. Environmental dominance in binary estimation at every blocklength yields a finite-code reliability--secrecy bound. These results explain why complete comparison prevents activation with antidegradable helpers whereas regularized comparison alone does not. Building on that qutrit example, we establish this stability for a range of noisy Werner--Holevo channels, determine sharp private-capacity and antidegradability thresholds, and compute exact complementary capacities in a nondegradable range. By contrast, we extend Pauli half-erasure activation to all $1/2\le p<1$ and establish the same range for a new four-level family. Reference-assisted measured-$χ^2$ witnesses show why these activating constructions lack complete comparison.
AI Summary
The paper asks how estimation-based channel comparisons certify zero private capacity beyond antidegradability, and why some zero-capacity channels activate with antidegradable helpers while others cannot. It connects binary minimum mean-square estimation, measured $\chi^2$ divergence, signed observable lifts, and entropic channel orders. A coherent-qubit integral representation proves that complete measured-$\chi^2$ ordering implies complete BKM-$\chi^2$, relative-entropy, and less-noisy ordering. A Hermitian-smoothed measured-$\chi^2$ comparison characterizes complete variance-dominating signed lifts; the converse uses coupled quadratic tests and minimax. Environmental MMSE dominance for every binary encoding at every blocklength yields the finite-code bound $(1-2\eta)^2\le2\delta$, relating decoding error $\eta$ and trace-distance secrecy error $\delta$. Complete anti-less-noisiness is also characterized by zero private capacity after pairing with every finite-dimensional half-erasure channel. For noisy qutrit Werner–Holevo channels $\Lambda_x$, private capacity vanishes exactly for $x\ge1/2$, whereas antidegradability begins at $x=4/7$. Their complements have exact capacities $P=Q=h_2(x)-(1-x)\log_2 3$ for $x\ge1/2$, including a nondegradable interval. These conclusions use inherited signed-lift certificates, symmetric extensions, covariance, and coherent-information concavity. For a previously constructed two-qubit Pauli channel, erasure-assisted activation extends to every $1/2\le p<1$, with rate at least $\ln2(1-p)^2/[216p(1+\sqrt3)]$. A continuous four-level ladder family activates throughout the same range. Transpose simulation certifies zero individual capacity; receiver support asymmetry supplies linear signal gain against quadratic environmental leakage. Open limitations include whether complete unsmoothed measured-$\chi^2$ ordering generally implies a signed lift, and whether every non-antidegradable transpose-antidegradable channel admits an antidegradable activator. Activation rates are lower bounds, not optimized capacities.
contributed to the mathematical proofs in this work, including proposing proof strategies, deriving intermediate results, and drafting and revising proof arguments. It also assisted with literature searches, numerical checks, exposition, figures, and \LaTeX{} production.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 40 pages