Research On Anti-frequency Sweeping Jamming Method For CW Doppler Fuze

With the electromagnetic environment of modern battlefield becomes more and more complex, the radio fuze faces serious interference. The CW(continuous wave) Doppler radio fuze is widely used in conventional ammunition weapon systems now, but the theoretical analysis and experimental results show that this kind of fuze’s ability of anti-frequency sweeping jamming is poor. In order to improve the anti-frequency sweeping jamming performance of the CW Doppler fuze, the thesis designed an anti-frequency sweeping jamming method for the CW Doppler fuze, which based on Fourier spectrum feature extraction of detector output signal and support vector machine classification and recognition. Experimental results showed that the method could achieve high classification accuracy, and it can effectively improve the anti- frequency sweeping jamming capability of the CW Doppler fuze.The thesis first analyzed the Fourier spectrum of the CW Doppler fuze’s detector output signal which under the effect of frequency sweeping jamming signal and target echo signal, improved the theoretical basis of feature extraction method that based on the peak points’ amplitude ratio of Fourier amplitude spectrum. Then the thesis compared binary-class Support Vector Machine and one-class Support Vector Machine, clarified that one-class Support Vector Machine had more advantage under the application background of the thesis. On the basis of algorithm analysis, the thesis used FPGA as the core and designed a signal acquisition and processing system, that is the implementation in hardware of the method.The hardware circuit included fuze detection signal conditioning circuit, starting signal conditioning circuit, FPGA minimum system, ADC acquisition circuit, serial communication circuit and power circuit; FPGA logic functions included serial communications, command analysis, FFT calculation, peak points searching, ratio calculation, classification decision function operation; in order to facilitate the data display and storage at the training sample collection stage, a PC control platform based on LabVIEW was also designed.Experiments of detector output signal classification and recognition for a certain type of CW Doppler fuze were done at the microwave chamber. The results showed that the hardware circuitry could reliably acquire signal, and the FPGA could complete the featureextraction based on the Fourier spectrum and the classification and recognition based on the Support Vector Machine. The increase in signal processing time was less than 0.66 ms, fully meet the requirements for signal processing time of the fuze.