Dynamical Quantum Tomography

TL;DR

This paper introduces a dynamical quantum tomography method where states are identified through time-evolved measurements, demonstrating efficient discrimination with minimal measurement outcomes under suitable dynamics and prior information.

Contribution

It presents a novel approach to quantum state discrimination using time evolution and fixed measurements, reducing the number of outcomes needed for accurate tomography.

Findings

Any two states in an n-dimensional system can be discriminated with n outcomes at n+1 time points.

Prior information restricts the set of states, allowing discrimination with fewer measurement outcomes.

Beyond unitary dynamics, the minimal two-outcome measurement setup suffices for state discrimination.

Abstract

We consider quantum state tomography with measurement procedures of the following type: First, we subject the quantum state we aim to identify to a know time evolution for a desired period of time. Afterwards we perform a measurement with a fixed measurement set-up. This procedure can then be repeated for other periods of time, the measurement set-up however remains unaltered. Given an $n$-dimensional system with suitable unitary dynamics, we show that any two states can be discriminated by performing a measurement with a set-up that has $n$ outcomes at $n+1$ points in time. Furthermore, we consider scenarios where prior information restricts the set of states to a subset of lower dimensionality. Given an $n$-dimensional system with suitable unitary dynamics and a semi-algebraic subset $\mathcal{R}$ of its state space, we show that any two states of the subset can be discriminated by performing a measurement with a set-up that has $n$ outcomes at $l$ steps of the time evolution if $(n-1)l\ge 2\dim\mathcal{R}$. In addition, by going beyond unitary dynamics, we show that one can in fact reduce to a set-up with the minimal number of two outcomes.