The new criterion to evaluate SAR jamming effect based on the statistical distinction

In this paper, a new criterion which is based on the statistical distinction to evaluate SAR jamming effect is proposed. The radar signal and jamming signal come from different signal source, and the statistical characteristics are not same, thereby the two signals are considered as independent. Here, the SAR images jammed can be look as the signal source mixed target signal with jamming signal, the jamming removal problem can also be described as blind signal separation problem. Using Independent Component Analysis (ICA) method, which is regarded as the statistical distinction tool of the mixture signal source (SAR images), estimates the orthogonal basis from the images. Moreover, we introduce the kurtosis as the criterion to classify the basis images and select threshold to decide which basis images span into target signal subspace and which basis images span jamming signal subspace. The SAR images jammed by Gaussian noise jamming and similar clutter jamming(SCJ) method are used as the input to the ICA algorithm respectively, the results provide that the effect on the jamming removing is degraded contrasted to the Gaussian noise jamming, because of the clutter jamming signal is similar to the target returned signal on the statistical characteristic.