Intelligent radar HRRP target recognition based on CNN-BERT model

Abstract Stable and reliable feature extraction is crucial for radar high-resolution range profile (HRRP) target recognition. Owing to the complex structure of HRRP data, existing methods fail achieve satisfactory performance. This study proposes a new deep learning model named convolutional neural network–bidirectional encoder representations from transformers (CNN-BERT), using spatio–temporal embedded in The token embedding module characterizes local spatial generates sequence features by embedding. BERT captures long-term temporal dependence among cells within through multi-head self-attention mechanism. Furthermore, novel cost function that simultaneously considers recognition rejection ability designed. Extensive experiments on measured data reveal superior performance proposed model.