Two methods for simulating non-Gaussian highly correlated sea clutter maps

Sea clutter features change significantly when measured by high resolution radars, at low grazing angles. The recorded data may be highly correlated and are characterized by long tail probability density functions, which are in the most cases very different compared to the classical Rayleigh distribution. In this paper we propose two methods for simulating this type of clutter. The former is based on the statistical model of the sea clutter, which can be seen as a combination of two random processes, which are Gamma and Rayleigh distributed. The second makes use of the 2D MA analysis of real sea clutter maps to generate synthetic data having very similar statistic and correlation properties.