Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:quant-ph/9912103 · Journal of Mathematical Physics

Stationary quantum source coding

Dénes Petz, Milán Mosonyi

Published Oct 1, 2001• 17 citations
arXivHTMLPDF

Abstract

In this article the quantum version of the source coding theorem is obtained for a completely ergodic source. This result extends Schumacher’s quantum noiseless coding theorem for memoryless sources. The control of the memory effects requires some earlier results of Hiai and Petz on high probability subspaces. Our result is equivalently considered as a compression theorem for noiseless stationary channels.

DOI
10.1063/1.1398335

Related papers

Stationary quantum source coding

Dėnes Petz, Milán Mosonyi

In this paper the quantum source coding theorem is obtained for a completely ergodic source. This results extends Shannon's classical theorem as well as Schumacher's quantum noiseless coding theorem for memoryless sources. The control of the memory effects requires earlier results of Hiai and Petz on high probability subspaces.

•Dec 23, 1999•24 citations

Data Compression for Quantum Code

Kwok Yip Szeto

The general scheme of data compression using the quantum noiseless coding theorem of Schumacher is dicussed for general quantum sources. When the Hilbert space of the quantum source is decomposable into orthogonal subspaces, one can first perform classical data compression before performing the quantum data compression for individual subspaces. For minimizing the resource in quantum coding, a general parameterization of the dimensions in the Jozsa-Schumacher quantum data compression scheme is presented.

arXiv:quant-ph/9607010•Jul 12, 1996•1 citation

A quantum version of Feinstein's Theorem and its application to channel coding

Nilanjana Datta, T. C. Dorlas

In this paper, a quantum version of Feinstein's theorem is developed. This is then used to give a completely self-contained proof of the direct channel coding theorem, for transmission of classical information through a quantum channel with Markovian correlated noise. Our proof does not rely on the Holevo-Schumacher-Westmoreland (HSW) theorem. In addition, for the case of memoryless channels, our method yields an alternative proof of the HSW Theorem

IEEE International Symposium on Information Theory (ISIT)•Jul 1, 2006•6 citations

On the reversible extraction of classical information from a quantum source

Howard Barnum, Patrick M. Hayden, Richard Jozsa +1 more

Consider a source ϵ of pure quantum states with von Neumann entropy S. By the quantum source coding theorem, arbitrarily long strings of signals may be encoded asymptotically into S qubits per signal (the Schumacher limit) in such a way that entire strings may be recovered with arbitrarily high fidelity. Suppose that classical storage is free while quantum storage is expensive and suppose that the states of ϵ do not fall into two or more orthogonal subspaces. We show that if ϵ can be compressed with arbitrarily high fidelity into A qubits per signal plus any amount of auxiliary classical storage, then A must still be at least as large as the Schumacher limit S of ϵ. Thus no part of the quantum information content of ϵ can be faithfully replaced by classical information. If the states do fall into orthogonal subspaces, then A may be less than S, but only by an amount not exceeding the amount of classical information specifying the subspace for a signal from the source.

arXiv:quant-ph/0011072•Proceedings of the Royal Society A Mathematical Physical and Engineering Sciences•Jul 8, 2001•25 citations

Experimental Demonstration of Quantum Source Coding

Yasuyoshi Mitsumori, John A. Vaccaro, Stephen Mark Barnett +4 more

We report an experimental demonstration of Schumacher's quantum noiseless coding theorem. Our experiment employs a sequence of single photons, each of which represents three qubits in terms of eight spatial and polarization modes. We initially prepare each photon in one of a set of eight nonorthogonal code word states corresponding to the value of a block of three binary letters. We use quantum coding to compress this quantum data into a two-qubit quantum channel and then uncompress the two-qubit channel to restore the original data with a fidelity approaching the theoretical limit.

arXiv:quant-ph/0304036•Physical Review Letters•Nov 19, 2003•15 citations

New Approaches on Source Coding for Quantum Stochastic Sources and Implementation of Quantum Fanout Gate

Rabins Wosti

The accurate computation of advanced quantum algorithms like Shor’s integer factorization, quantum phase estimation (QPE), and the quantum Fourier transform (QFT) requires quantum circuits of considerable size and depth. It is difficult to achieve reliable computation with deep quantum circuits due to the limited coherence times of the current noisy quantum devices. The quantum fanout gate is known to be a powerful primitive for reducing the depth of many quantum circuits (Høyer and Špalek 2003; Gottesman and Chuang 1999). Shallow or constant-depth quantum circuits are desirable for both near-term and fault-tolerant quantum computations as they reduce noise and allow faster execution of quantum algorithms, potentially skirting the effects of short coherence times. In this work, we show new approaches towards implementation of quantum fanout gate. In particular, we show that by analogue time-evolving the quantum systems according to two well-studied Hamiltonians, namely, quantum Ising and quantum Heisenberg models, we can implement the quantum fanout operator using constantly many layers of digital quantum gates. Important foundations to the area of quantum encoding were provided by Schumacher who proved the quantum analog of Shannon’s noiseless coding theorem for an independent and identically distributed (i.i.d.) quantum source, (Schumacher 1995). In this work, we show a lossless, variable-length block encoding scheme of quantum information emitted from a completely general stochastic quantum source, and it is encoded into the Fock space. While doing so, we extend the notion of uniquely decodable (or completely lossless) quantum codes to be used for quantum block data compression. As our main result, for a fixed mlmany pure states emitted by a given quantum stochastic source, we derive the optimal lower bound of the average codeword length over a subset of uniquely decodable quantum codes called “special block codes,” which are applied to encode the pure states of m many blocks each of block size l. Additionally, we show that for quantum stationary sources in particular, the optimal lower bound of the average codeword length per symbol computed over a subset of special block codes called “constrained special block codes” equals the von Neumann entropy rate of the source for an asymptotically long block size.

•Oct 1, 2025•0 citations
In this article the quantum version of the source coding theorem is obtained for a completely ergodic source. This result extends Schumacher’s quantum noiseless coding theorem for memoryless sources. The control of the memory effects requires some earlier results of Hiai and Petz on high probability subspaces. Our result is equivalently considered as a compression theorem for noiseless stationary channels.
The general scheme of data compression using the quantum noiseless coding theorem of Schumacher is dicussed for general quantum sources. When the Hilbert space of the quantum source is decomposable into orthogonal subspaces, one can first perform classical data compression before performing the quantum data compression for individual subspaces. For minimizing the resource in quantum coding, a general parameterization of the dimensions in the Jozsa-Schumacher quantum data compression scheme is presented.
Consider a source ϵ of pure quantum states with von Neumann entropy S. By the quantum source coding theorem, arbitrarily long strings of signals may be encoded asymptotically into S qubits per signal (the Schumacher limit) in such a way that entire strings may be recovered with arbitrarily high fidelity. Suppose that classical storage is free while quantum storage is expensive and suppose that the states of ϵ do not fall into two or more orthogonal subspaces. We show that if ϵ can be compressed with arbitrarily high fidelity into A qubits per signal plus any amount of auxiliary classical storage, then A must still be at least as large as the Schumacher limit S of ϵ. Thus no part of the quantum information content of ϵ can be faithfully replaced by classical information. If the states do fall into orthogonal subspaces, then A may be less than S, but only by an amount not exceeding the amount of classical information specifying the subspace for a signal from the source.
We report an experimental demonstration of Schumacher's quantum noiseless coding theorem. Our experiment employs a sequence of single photons, each of which represents three qubits in terms of eight spatial and polarization modes. We initially prepare each photon in one of a set of eight nonorthogonal code word states corresponding to the value of a block of three binary letters. We use quantum coding to compress this quantum data into a two-qubit quantum channel and then uncompress the two-qubit channel to restore the original data with a fidelity approaching the theoretical limit.
The accurate computation of advanced quantum algorithms like Shor’s integer factorization, quantum phase estimation (QPE), and the quantum Fourier transform (QFT) requires quantum circuits of considerable size and depth. It is difficult to achieve reliable computation with deep quantum circuits due to the limited coherence times of the current noisy quantum devices. The quantum fanout gate is known to be a powerful primitive for reducing the depth of many quantum circuits (Høyer and Špalek 2003; Gottesman and Chuang 1999). Shallow or constant-depth quantum circuits are desirable for both near-term and fault-tolerant quantum computations as they reduce noise and allow faster execution of quantum algorithms, potentially skirting the effects of short coherence times. In this work, we show new approaches towards implementation of quantum fanout gate. In particular, we show that by analogue time-evolving the quantum systems according to two well-studied Hamiltonians, namely, quantum Ising and quantum Heisenberg models, we can implement the quantum fanout operator using constantly many layers of digital quantum gates. Important foundations to the area of quantum encoding were provided by Schumacher who proved the quantum analog of Shannon’s noiseless coding theorem for an independent and identically distributed (i.i.d.) quantum source, (Schumacher 1995). In this work, we show a lossless, variable-length block encoding scheme of quantum information emitted from a completely general stochastic quantum source, and it is encoded into the Fock space. While doing so, we extend the notion of uniquely decodable (or completely lossless) quantum codes to be used for quantum block data compression. As our main result, for a fixed mlmany pure states emitted by a given quantum stochastic source, we derive the optimal lower bound of the average codeword length over a subset of uniquely decodable quantum codes called “special block codes,” which are applied to encode the pure states of m many blocks each of block size l. Additionally, we show that for quantum stationary sources in particular, the optimal lower bound of the average codeword length per symbol computed over a subset of special block codes called “constrained special block codes” equals the von Neumann entropy rate of the source for an asymptotically long block size.