FDDphase: A Package For Statistical Analysis of the Phase from Polarimetric Data

Abstract

Polarimetric Synthetic Aperture Radar data provide a wealth of information. A complete single-frequency observation of this kind is a complex matrix with nine degrees of freedom: three intensities and three complex covariances. There are several ways of extracting valuable information from these observations, among them the use of statistical models. In particular, the difference between the phases of the complex covariances is important in interferometric applications. In this article we survey statistical models for this kind of circular data. We present both physically-derived and empirical distributions, we highlight their relationships and present estimators for their parameters in an unified manner. We present a package for the R platform for statistical analysis of phase data under univariate physical and empirical models, their estimators, and their relationships.