On optimal filtering of measured Mueller matrices

TL;DR

This paper introduces a convex decomposition framework for Mueller matrices, enabling optimal filtering of noise in polarimetric measurements by characterizing the system's randomness.

Contribution

It proposes a novel convex decomposition method for Mueller matrices that aids in noise filtering and system characterization in polarimetry.

Findings

Convex decomposition effectively characterizes polarimetric randomness.

Framework provides criteria for optimal noise filtering.

Applicable to experimental polarimetric data.

Abstract

While any two-dimensional mixed state of polarization of light can be represented by a combination of a pure state and a fully random state, any Mueller matrix can be represented by a convex combination of a pure component and three additional components whose randomness is scaled in a proper and objective way. Such characteristic decomposition constitutes the appropriate framework for the characterization of the polarimetric randomness of the system represented by a given Mueller matrix, and provides criteria for the optimal filtering of noise in experimental polarimetry.