Measurement and Characterization of Entropy and Degree of Polarization of Weather Radar Targets

Abstract

To date, few polarimetric weather radars have exhibited the capability to measure full scattering matrices. In contrast, in the synthetic aperture radar (SAR) community, considerable experience has been gained in dealing with complete scattering matrices and their statistical behavior. This paper aims to place weather radar parameters in a wider context in order to exploit more general concepts like target decomposition theorems and polarization basis transformations. Entropy, which is a fully polarimetric variable derived from the Cloude-Pottier decomposition, and the degree of polarization, which is derived from Wolf's coherence matrix, are the subject of this paper. The theoretical analysis carried out in the first part is checked against fully polarimetric data from POLDIRAD, which is the German Aerospace Center research weather radar. The entropy and the degree of polarization are compared with the copolar correlation coefficient in order to understand whether they can add value to radar meteorological investigations. Because the degree of polarization is available to conventional dual-polarization coherent systems, it is important to assess its potential for operational use.

Eigenvalue signal processing for phased-array weather radar polarimetry: Removing the bias induced by antenna coherent cross-channel coupling

“…On the other hand, all variables presented in this paper are exact replacements, that is, the eigenvalue variables derived from the Coherency matrix do correspond exactly to the unbiased standard polarimetric radar variables. Entropy H and the copolar correlation coefficient ρ hv do have a similar physical meaning and generally do display the same discrimination capabilities [5]. Entropy is a measure of the diversity of scattering matrices that form the Covariance matrix, the copolar correlation coefficient is a measure of the diversity of H to V axis ratios of the scatterers.…”

“…At ATSR mode however, both the copolar correlation coefficient ρ hv and the specific differential phase KDP are not significantly affected by antenna cross-channel coupling [4]. For completeness' sake, we mention that an eigenvalue-derived proxy for the copolar correlation coefficient ρ hv is scattering entropy H [5]. This case however is profoundly different: contrary to all variables treated in the present paper, which are derived from the eigenvalues of the Coherency matrices at H and V transmit (upper left and lower right minors of the Covariance matrix), scattering entropy is derived from the eigenvalues of the 3x3 Covariance matrix.…”

Zenith/Nadir Pointing mm-Wave Radars: Linear or Circular Polarization?

“…Under the assumption of azimuthal symmetry, the Mueller matrix can in general be written in terms of Huynen parameters as [3], [37] …”

Geometric Polarimetry—Part II: The Antenna Height Spinor and the Bistatic Scattering Matrix

“…Although multilook averaging in SAR [1] for natural targets and multisampling in weather radar [2] involve partially coherent averaging techniques there is now more than ever a need to understand fully both real and abstract geometric relationships in coherent scattering. The use of basis or geometric transformations, in solving polarimetric problems continues to feature in the literature [3], [4] [5]. In recent years there has been an increase in interest in polarimetric bistatic scattering as technical capabilities for multi-platform coherent systems (e.g.…”