Anti-jamming Method of FM Fuze Based on Deep Learning

Abstract In response to the challenges posed by sweeping jamming and Digital Radio Frequency Memory (DRFM) jamming in Frequency Modulation Continuous Wave (FMCW) fuzes, a jamming mitigation method based on deep learning method is proposed in this paper. Initially, the target beat frequency signal data affected by sweeping jamming and DRFM jamming is gathered and analysed using the Transformer method. A comparison is made between the classification and recognition outcomes of Transformer model and Support Vector Machine (SVM) methods to identify the optimal model. Subsequently, the feature information of the echo signal under interference is extracted using Empirical Mode Decomposition (EMD) decomposition, and the target features are reconstructed. Experimental results utilizing simulated data confirm that the proposed method achieves a classification accuracy of 92% and ensures that the fuze detonation distance remains within the pre-set 9 m range.