Radar Jamming Effectiveness Evaluation Method Based on Few-Shot Learning

This paper proposes a radar jamming effectiveness evaluation method based on small-sample learning, aiming to address the weak generalization and robustness of traditional evaluation methods in data-scarce scenarios. By combining suppression and deception jamming parameters for different preset radar waveform types, a small-sample task training set is constructed, and the evaluation model is trained using the Reptile meta-learning algorithm. Experiments demonstrate that this method achieves high-precision jamming effectiveness evaluation even with a small number of samples, significantly improving the model's adaptability in dynamic battlefield environments.