A neural network approach to weibull distributed sea clutter parameter’s estimation

Abstract

The main problem faced by sea radars is the elimination of an undesirable signal that appears mixed with target information: sea clutter. One of the most popular probability distributions in clutter modelling is the Weibull distribution. Helpful in efficient detectors’ design, a system able to recognize the Weibull shape parameter knowing a priori that the mean of the distribution is equal to one is proposed. The solution achieves a more precise estimation than the traditional method of moments, and it’s appropriate for real time operating conditions as it is based on a neural network approximation.