Sea clutter power reduction in radar measurement systems by feedforward multilayer perceptrons with medium input data integration rate

In radio measurement systems, the backscatter from what is not a target, i.e. the clutter, is usually not desired. So, these systems try to incorporate clutter reduction techniques as efficient as possible. In this way, different signal processing techniques can be used. The case of study presented in this paper shows how to reduce the level of sea clutter measurements in a marine radar. Due to linear linear signal processing is not suitable in these cases, nonlinear signal processing is used, which is achieved by neural networks. In this way, 7 cells to evaluate the output of each cell under test of a radar image (medium input integration rate) are selected. The processed radar images show very promising results from a subjective point of view. On the other hand, objective measurements are used to analyzed the system performance. Those measurements are based on the mean square error and the clutter and target powers at the input and output of the proposed clutter reduction system. Minimum and mean clutter reduction power rates of 7 dB and 10 dB are achieved, respectively.