State estimation of long-range correlated non-equilibrium systems: media estimation

TL;DR

This paper investigates how to estimate the structure of non-equilibrium media with long-range correlations using limited spatial data, comparing linear and nonlinear estimators for accuracy and bias.

Contribution

It introduces novel statistical estimators for media structure from scattered data in non-equilibrium systems, highlighting their advantages and limitations.

Findings

Linear estimator provides unbiased density profiles.

Nonlinear maximum likelihood estimator accurately captures media structure.

Estimators reveal long-range correlations in non-equilibrium systems.

Abstract

Non-equilibrium systems have long-ranged spatial correlations even far away from critical points. This implies that the likelihoods of spatial steady state profiles of physical observables are nonlocal functionals. In this letter, it is shown that these properties are essential to a successful analysis of a functional level inverse problem, in which a macroscopic non-equilibrium fluctuation field is estimated from limited but spatially scattered information. To exemplify this, we dilute an out-of-equilibrium fluid flowing through random media with a marker, which can be observed in an experiment. We see that the hidden variables describing the random environment result in spatial long-range correlations in the marker signal. Two types of statistical estimators for the structure of the underlying media are then constructed: a linear estimator provides unbiased and asymptotically precise information on the particle density profiles, but yields negative estimates for the effective resistances of the media in some cases. A nonlinear, maximum likelihood estimator, on the other hand, results in a faithful media structure, but has a small bias. These two approaches complement each other. Finally, estimation of non-equilibrium fluctuation fields evolving in time is discussed.