Quantum probes for the spectral properties of a classical environment

Abstract: We address the use of simple quantum probes for the spectral characterization of classical noisy environments. In our scheme a qubit interacts with a classical stochastic field describing environmental noise and is then measured after a given interaction time in order to estimate the characteristic parameters of the noise. In particular, we address estimation of the spectral parameters of two relevant kinds of non-Gaussian noise: random telegraph noise with Lorentzian spectrum and colored noise with $1/f^{\alpha}$ spectrum. We discuss the estimation precision achievable by quantum probes by analyzing in details the behavior of the quantum Fisher information (QFI) as a function of the noise and the other parameters. We found that population measurement on the qubit is optimal for noise estimation and also evaluate the optimal interaction times for the quantum probe, i.e. the values maximizing the QFI and the quantum signal-to-noise ratio. For random telegraph noise the QFI is inversely proportional to the square of the switching rate, meaning that the quantum signal-to-noise ratio is constant and thus the switching rate may be uniformly estimated with the same precision in its whole range of variation. For colored noise, the precision achievable in the estimation of ``color'', i.e. of the exponent $\alpha$, strongly depends on the structure of the environment, i.e. on the number of fluctuators describing the classical environment. For an environment modeled by a single random fluctuator estimation is more precise for pink noise, i.e. for $\alpha=1$, whereas by increasing the number of fluctuators, the QFI is has two local maxima, with the largest one drifting towards $\alpha=2$, i.e. brown noise.