Jamming Style Selection for Small Sample Radar Jamming Rule Base

Abstract

Aiming at the poor effect of traditional electronic warfare (EW) jamming style selection on modern multifunctional radar, a jamming style selection model based on jamming rule base and the support vector machine (SVM) jamming style selection method suitable for the model is proposed to improve the jamming style selection correct rate and reduce the jamming style selection time. For air-to-air scene of airborne multifunction fire control radar, the pulse repetition frequency (PRF), carrier frequency (CF), pulse width (PW) and intra-pulse box dimension (BoxD), spectral resemblance coefficient (Cr) characteristics of radar signal are extracted and a small sample jamming rule base is constructed. The SVM theory is used to select the jamming style, and the inner product of the input space is transformed into the inner product of the feature space by using the nonlinear mapping radial basis function (RBF). The feature space hyperplane is established and the input is labeled as the corresponding jamming style label. Jamming style is selected shortly and effectively. The experimental results show that compared with the traditional template matching (TM) method, the proposed method can effectively improve the accuracy and real-time performance of jamming style selection. The practical problem of small sample and non-linear interference style selection are solved for its excellent learning and generalization ability.