Frequency modulation fuze target recognition method based on sparse representation of wavelet packet dictionary

Aiming at the situation that the complex battlefield electromagnetic environment jamming with the radio fuze, which causes early explosion and weakens the striking efficiency, a frequency modulation target recognition method based on sparse representation of wavelet packet dictionary is proposed for the sweep frequency jamming, which is the most threatening jamming to the frequency modulation fuze. The sparse dictionary is constructed by using the three-layer wavelet packet entropy feature as the atom. The output signal of the fuze detection end is sparsely decomposed in the wavelet packet sparse dictionary, and the sparse representation coefficient is sparsely reconstructed. The classification of the fuze target signal is determined by the principle of minimum reconstruction error. The experimental verification of the measured data of the frequency modulation radio fuze prototype shows that the accuracy of target recognition reaches 98.72%, and the accuracy of interference recognition reaches 97.94%. This shows that the proposed method can effectively recognize the target signal of frequency modulation fuze and several typical sweep frequency jamming signals.