Anti-jamming method based on multi-instance multi-label learning

This paper analyses jamming recognition and countermeasure based on MIML (multi-instance multi-label learning). We first use the kernel clustering algorithm for signal sorting to determine the interference signal and its frequency. Based on the MIML framework, it extracts more efficient and comprehensive time-frequency distribution characteristics in the interference signal in the form of compound signal. After recognizing the modulation mode of interference, the particle swarm optimization algorithm is used to search for the waveform with the lowest correlation with the interference waveform. By adjusting the waveform parameters, the waveform in the working range is as uncorrelated with the interference as possible, and the interference is eliminated based on waveform optimization. Finally, simulations are carried out to examine the performance of the proposed method.