Multiresolution Jamming Recognition with Few-shot Learning

Abstract

The few-shot learning method, based on multiresolution feature extraction, is proposed for the recognition of jamming signals, including range gate pull-off (RGPO), velocity gate pull-off (VGPO), and range-velocity gate pull-off (RVGPO) jamming signals. The general jamming recognition technology is mature, but a large number of training samples are needed. In this paper, the prototype network is used for few-shot learning that taking features extracted from wavelet decomposition and empirical mode decomposition (EMD). It is shown that the proposed method has a good recognition effect under the condition of few samples.