Dictionary Learning and Waveform Design for Dense False Target Jamming Suppression

─ For linear frequency modulation (LFM) pulse radars, dense false targets generated by new system jamming seriously damage the performance of such radar systems. In order to avoid the influence of dense false target jamming, an anti-jamming strategy combining waveform design and sparse decomposition are proposed. Specifically, the radar system transmits a random pulse initial phase (RPIP) signal, and uses peak detection method to detect the deception jamming. The phase distribution of the RPIP signal is partially randomly perturbed for a jamming, and we use optimization algorithm to design a phase perturbed LFM (PPLFM) signal with good autocorrelation characteristics. Using the correlation function of the designed signal, the target sample set and the jamming sample set are constructed, and the target echo and the jamming signal are separated using designed dictionary learning method to achieve suppression of dense false target jamming and range side-lobes. The effectiveness of the proposed method is verified by numerical simulation, and the results proved that this proposed method maintains good anti-jamming performance under low signal-tonoise ratio (SNR). Index Terms ─ Anti-jamming, dense false target jamming, dictionary learning, jamming detection, waveform design.