How state preparation can affect a quantum experiment: Quantum process tomography for open systems

TL;DR

This paper investigates how different state preparation methods in quantum process tomography affect the resulting process maps, revealing that preparation procedures can lead to linear or non-linear descriptions of quantum processes.

Contribution

It introduces a new process tomography method for measurement-based preparations and analyzes how preparation procedures influence process map linearity.

Findings

Stochastic preparation yields linear process maps.

Measurement-based preparation can produce non-linear, bi-linear process maps.

A verification protocol differentiates linear from bi-linear processes.

Abstract

We study the effects of preparation of input states in a quantum tomography experiment. We show that maps arising from a quantum process tomography experiment (called process maps) differ from the well know dynamical maps. The difference between the two is due to the preparation procedure that is necessary for any quantum experiment. We study two preparation procedures, stochastic preparation and preparation by measurements. The stochastic preparation procedure yields process maps that are linear, while the preparations using von Neumann measurements lead to non-linear processes, and can only be consistently described by a bi-linear process map. A new process tomography recipe is derived for preparation by measurement for qubits. The difference between the two methods is analyzed in terms of a quantum process tomography experiment. A verification protocol is proposed to differentiate between linear processes and bi-linear processes. We also emphasize the preparation procedure will have a non-trivial effect for any quantum experiment in which the system of interest interacts with its environment.