An empirical model for the shape parameter of K distribution in radar sea clutter at low grazing angles

Accurately predicting the shape parameter of the K-distribution under specific radar and environmental conditions plays an important role in radar sea clutter simulation and constant false alarm rate (CFAR) detection. However, most existing empirical models primarily rely on radar grazing angle and resolution area, with their weight parameters varying across sea states, thereby limiting their generalizability under diverse environmental conditions. In this paper, an improved empirical model for the shape parameter of the K distribution is proposed, expressing it as a function of the normalized radar cross section (NRCS) of sea surfaces and the area of the radar resolution cell. The introduction of NRCS provides a more explicit link between the shape parameter and environmental factors. Consequently, the weight parameters of the model can be estimated using dataset aggregated across all sea states and wind direction, eliminating the need for sea-state-specific fitting. Moreover, by leveraging an expanded dataset obtained through range resolution reduction processing, it is found that an exponential function provides the closest fit for the relationship between the shape parameter and radar resolution area, leading to appropriate corrections in the improved model. Experimental results on measured sea clutter data demonstrate that the proposed model achieves much higher prediction accuracy compared to existing empirical models.