Polarimetric modeling and parameter estimation with electromagnetic vector sensors

We analyze two polarimetric models in which the complete electromagnetic field is measured. The first model considers estimation of the polarimetric response of a surface by measuring the reflections of actively generated plane waves. The second model considers estimation of the polarization of passively generated plane waves. In both models we provide a natural parametrization of the distribution of the received electromagnetic signal. Using the Cramer-Rao bound on the variance of unbiased estimates of the parameters, we show that this parametrization, which has a simple geometrical interpretation, also has good statistical properties. We fit these models into a common framework and compare them.