Entanglement Cost of Quantum Channels

TL;DR

This paper derives a formula for the entanglement cost of quantum channels, linking it to the entanglement of formation and providing insights into quantum communication and security.

Contribution

It introduces a regularised optimization formula for the entanglement cost of quantum channels, extending the entanglement of formation concept and connecting to the quantum reverse Shannon theorem.

Findings

The entanglement cost formula is analogous to the entanglement of formation for states.

Security in the noisy-storage model can be enhanced using the entanglement cost of channels.

Sending quantum information above the entanglement cost rate results in exponentially small fidelity.

Abstract

The entanglement cost of a quantum channel is the minimal rate at which entanglement (between sender and receiver) is needed in order to simulate many copies of a quantum channel in the presence of free classical communication. In this paper we show how to express this quantity as a regularised optimisation of the entanglement formation over states that can be generated between sender and receiver. Our formula is the channel analog of a well-known formula for the entanglement cost of quantum states in terms of the entanglement of formation; and shares a similar relation to the recently shattered hope for additivity. The entanglement cost of a quantum channel can be seen as the analog of the quantum reverse Shannon theorem in the case where free classical communication is allowed. The techniques used in the proof of our result are then also inspired by a recent proof of the quantum reverse Shannon theorem and feature the one-shot formalism for quantum information theory, the post-selection technique for quantum channels as well as Sion's minimax theorem. We discuss two applications of our result. First, we are able to link the security in the noisy-storage model to a problem of sending quantum rather than classical information through the adversary's storage device. This not only improves the range of parameters where security can be shown, but also allows us to prove security for storage devices for which no results were known before. Second, our result has consequences for the study of the strong converse quantum capacity. Here, we show that any coding scheme that sends quantum information through a quantum channel at a rate larger than the entanglement cost of the channel has an exponentially small fidelity.