Statistical Analysis of Real Polarimetric Clutter Data at Different Range Resolution., Part 1: Average Behavior

摘要

This report results from a contract tasking University of Pisa as follows: The research is divided in two phases. In the first phase we consider modeling and estimating the texture time/spatial variation in high resolution non-Gaussian clutter from RAR and SAR. We will investigate as well non- stationary and possible non-Gaussianity of the speckle component. The speckle distribution will be analyzed by both fitting to the data non-Gaussian distributions accounting for speckle non-Gaussianity, and by direct analysis of speckle through estimation of the I/Q ratio statistics (which decouples from the texture effects). The analyses of high resolution RAR clutter will be performed by processing complex clutter data collected in November 1993 and in winter 1998 with the high resolution IPIX radar, provided by McMaster University. The analyses of high resolution SAR clutter will be performed by processing clutter data (single-look complex) collected at Kohala Coast Hawaii, on 1906 with the high resolution AIRSAR airborne radar platform provided by NASA/Jet Propulsion Laboratory. An accurate characterization of the high resolution radar clutter allows a more adequate design of the radar detector of targets. This detector needs to be adaptive to the data and should have the ability to comply with the non-homogeneous nature of the clutter interference. This aspect of the problem will be discussed in a preliminary fashion in the second phase of this study. A more detailed analysis of the problem, which encompasses the design of the detector and the evaluation of its performance in operationally meaningful environment, will be tackled in a future phase of the study to be separately funded after this research.