Research on Parameters Predictive Methods for Sea Clutter Model

This paper utilizes the equivalence relationship of clutter fluctuation index provided by various models to establish the correlation between shape parameters of the Compound Gaussian amplitude distributions models. Moreover, empirical prediction methods for the shape parameters of different models are derived based on existing empirical formulas for the shape parameters of the K-distribution. The clutter power is described using a mean backscatter model, and a prediction method for the scale parameter is provided based on the relationship between shape parameters, scale parameters, and clutter power. By applying the aforementioned methods, this paper predicts the shape and scale parameters of different models based on the scene parameters of empirical data, and goodness-of-fit (GoF) tests are conducted to evaluate the fit between the amplitude distributions of different models and the empirical data. Experimental results demonstrate that applying the predicted shape and scale parameters using the methods proposed in this paper yields amplitude distributions of clutter that closely resemble the empirical data, and the fitting performance is influenced by the backscatter models. In particular, the Pareto amplitude distribution with the Naval Research Laboratory (NRL) and Technology Service Corporation (TSC) models outperforms other models in most cases. It can be concluded that the prediction of model parameters in this paper is an effective and superior method.