Introduction

What is the advantage of polarimetric SAR data and polarimetric techniques over the conventional SAR? Conventional SAR operate with a fixed single-polarization antenna for transmission and reception of radio frequency signals. In this way only a single scattering coefficient is measured for a specific combination of transmitted and received polarization, which is proportional to the received backscattered power at a particular combination of linear polarization (HH, HV, VH or VV). Due to this implementation the vectorial nature of the scattered wave is ignored and the additional information about the scattering process contained in the polarization properties of the signal is lost. To ensure that all the information in the scattered wave is retained, the polarization of the scattered wave must be measured through a vector measurement process. For a SAR system that coherently transmits and receives both signals from an an orthogonally polarized antenna pair (e.g. horizontal and vertical polarized), the elements of the resulting scattering matrix , the [S] - matrix can be measured. The [S] - matrix is therefore, a 2x2 complex scattering matrix yielding an eight dimensional measurement space. For reciprocal targets (where, for the backscattering case, $S_{hv} \equiv S_{vh}$, due to the reciprocity theorem), this space is compressed to five dimensions: three amplitudes (|HH|, |HV|, and |VV|); and two phase measurements, (co-polar: HH-VV, and cross-polar: HH-HV). It is the phase measurements which are distinctive of the technology, and have allowed the development of techniques to synthesize polarization measurements for any point on the Poincare sphere. This increases the measurement space of polarimetry far beyond the fivefold increase implied by the scattering matrix. When measuring all the four signals of the scattering matrix it is possible to synthesize the signal at any combination of transmit and receive polarizations. (i.e., any valid combination of ellipticity and orientation angles). In General, the scattering matrix is measured in the H, V basis, constituted by the HH, HV, VH, and VV signals. From this [S] - matrix the, for instance, the RL signal can be synthesized. Here H, V, R, and L means horizontal, vertical, right-handed circular, and left-handed circular polarization, respectively, and the first letter indicates the receive polarization while the second letter indicates the transmit polarization. Also the RR, LR, and LL signals can be synthesized and so the scattering matrix in another basis, namely the circular one, can be computed. Systems like the space borne SIR-C/X-SAR or the airborne Danish EMISAR and the German E-SAR systems provide such fully polarimetric data.

Fully polarimetric SAR data can be decomposed according to certain sets of targets with distinct scattering mechanisms, e.g.surface, diplane, 45$^o$ tilted diplane or sphere, diplane, helix. In these decompositions the polarimetric SAR data are expressed as a superposition of the afore mentioned three signal components/scattering mechanism contributions. Another useful decomposition is an eigenvalue and eigenvector based decomposition [Cloude99a]. The advantage of fully polarimetric SAR data is due to the fact that by evaluating the scattering matrix, various scattering mechanisms and target properties can be identified. In this treatment the basics of SAR polarimetry will be outlined and discussed. In the first part we will focus on the polarization and representation of plane waves. In the second part the description of scatteres will be discussed and various decomposition approaches for the analysis of the scatterers will be outlined. Finally, in the last part we will give some examples of the application of polarimetric techniques with respect to the interpretation of the fully polarimetric SAR data for various purposes.