A MODEL FOR SPATIALLY CORRELATED RADAR CLUTTER

On the basis of recent empirical evidence, a model for spatially correlated radar clutter is developed. In this model, echo amplitudes from two different resolution cells are described statistically by the correlated K-distribution. Explicit expressions for the joint probability density function and joint moments of this distribution are developed. The problem of minimum mean-square error (MMSE) clutter estimation is addressed where the clutter level in a given resolution cell is estimated from the clutter level in a different cell by minimizing the mean square error. Finally, the role of spatial correlation and resolution cell size on the overall characteristics of echo are considered. Results show that clutter in this class may be characterized by a family of K-distributions whose parameters are determined by the resolution cell area and the spatial autocorrelation function. These results are in agreement with reported measurements.