Dephasing superchannels

TL;DR

This paper introduces dephasing superchannels that selectively diminish quantum coherence in channels, characterizes their mathematical structure, physical realizations, and effects on channel distinguishability and coherence measures.

Contribution

It defines and analyzes dephasing superchannels, showing their mathematical form, physical implementations, and impact on quantum coherence and channel distinguishability.

Findings

Dephasing superchannels form a subclass of Schur-product supermaps.

Memory effects are significant for systems with dimension greater than 2.

Coherence generating power is monotone under dephasing superchannels.

Abstract

We characterise a class of environmental noises that decrease coherent properties of quantum channels by introducing and analysing the properties of dephasing superchannels. These are defined as superchannels that affect only non-classical properties of a quantum channel $\mathcal{E}$, i.e., they leave invariant the transition probabilities induced by $\mathcal{E}$ in the distinguished basis. We prove that such superchannels $\Xi_C$ form a particular subclass of Schur-product supermaps that act on the Jamiolkowski state $J(\mathcal{E})$ of a channel $\mathcal{E}$ via a Schur product, $J'=J\circ C$. We also find physical realizations of general $\Xi_C$ through a pre- and post-processing employing dephasing channels with memory, and show that memory plays a non-trivial role for quantum systems of dimension $d>2$. Moreover, we prove that coherence generating power of a general quantum channel is a monotone under dephasing superchannels. Finally, we analyse the effect dephasing noise can have on a quantum channel $\mathcal{E}$ by investigating the number of distinguishable channels that $\mathcal{E}$ can be mapped to by a family of dephasing superchannels. More precisely, we upper bound this number in terms of hypothesis testing channel divergence between $\mathcal{E}$ and its fully dephased version, and also relate it to the robustness of coherence of $\mathcal{E}$.