arXiv:2610.06840
A Spectrum-Based Converse for Quantum State Discrimination and Its Applications to Classical-Quantum Channel Coding
Tam{á}s Havas, Hsuan-Yin Lin, Eirik Rosnes
Abstract
We investigate converse bounds on the average decoding error probability in finite-blocklength classical-quantum channel coding. We first present a lower bound for multiple quantum hypothesis testing in terms of pairwise trace distances and derive a corresponding fidelity bound. We then obtain a spectrum-based converse that depends only on the a priori probabilities and spectra of the states. We show that the converse bound remains tight for the quantum depolarizing channel under suitable conditions. For codes with product-state outputs, this converse takes an explicit form involving products of output-state eigenvalues. We apply it to binary codes over the quantum amplitude damping channel using the input states $\vert+\rangle$ and $\vert-\rangle$. For this setting, we also discuss a normal approximation to the spectrum-based converse in the large blocklength regime. In all numerical examples considered, the spectrum-based converse is tighter than the other converse bounds at low noise levels.
AI Summary
The paper studies measurement-independent lower bounds on average decoding error in finite-blocklength classical–quantum channel coding, viewed as multiple quantum hypothesis testing. Its main question is what converse can be obtained using only the prior probabilities and individual output-state spectra, without eigenvectors or pairwise state relations. For $M$ states $\rho_i$ with priors $p_i$, it derives $P_e\geq 1-1/M-(1/M)\sum_{i<j}\|p_i\rho_i-p_j\rho_j\|_1$, which dominates the quoted Qiu bound, and converts this into a fidelity bound. The central spectral converse states that, in dimension $N$, optimal success probability is at most the sum of the $N$ largest entries in the pooled weighted spectra $\{p_i\lambda_j(\rho_i)\}$. It gives sufficient conditions for equality through spectral projectors forming a POVM, relates tightness to quasi-perfect codes, and discusses preservation of tightness under depolarizing noise under suitable conditions. For product-state codewords, the bound uses products of single-use eigenvalues and is related to sphere packing for an auxiliary classical channel. The paper reports general computational cost $O(MN^3)$, versus $O(M^2N^3)$ for the compared fidelity bounds, with further reductions for isospectral symbols. The stated application uses binary codes with $|+\rangle,|-\rangle$ inputs to the amplitude-damping channel. The abstract reports tighter bounds at low noise in the numerical examples and discusses a large-blocklength normal approximation. Tightness is conditional rather than general. The supplied text ends before those numerical results, the approximation, and the proof appendices.
Error found: The depolarization corollary is false for arbitrary priors: initially tight discrimination of orthogonal pure qubit states with priors (0.9, 0.1) yields optimal success 0.9 after depolarization with p=1/2, whereas its claimed preserved-index formula gives 0.75.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.