SVM-based continuous wave Doppler radio fuse frequency-swept jamming prevention method

The invention relates to an SVM-based continuous wave Doppler radio fuse frequency-swept jamming prevention method which comprises the following steps: step 1, acquiring a fuse detection signal, and performing fast Fourier transform on the sampled data; step 2, searching three maximum peak points of a Fourier amplitude spectrum; step 3, respectively solving the ratio of the amplitudes of the three maximum peak points; step 4, constructing a three-dimensional feature vector by using the Fourier amplitude spectrum peak value ratio so as to serve as an input training classifier of the SVM, thereby obtaining a classification decision function; and step 5, acquiring the features of the Fourier amplitude spectrum peak value ratio of an unknown fuse detection signal, inputting the features into the SVM, and classifying and identifying the signal by virtue of the classification decision function value. According to the method disclosed by the invention, the operating features of a frequency-swept jamming machine are fully utilized, the fuse detection signal under the actions of a target echo signal and a frequency-swept jamming signal can be accurately identified, and the frequency-swept jamming resistance of the fuse can be effectively improved.