Research on Individual Identification of Radar Emitters Based on Transformer Neural Networks

作者：Xiang Chen, Bi-Sheng Pan, Yu Chen · 发表于：International Conference on Speech Technology and Human-Computer Dialogue · 年份：2025 · DOI：10.1109/icct67417.2025.11374074

This paper proposes a radar emitter individual identification scheme based on Transformer neural networks. Leveraging the unique advantages of the multi-head attention mechanism, cross-attention mechanism, and self-attention mechanism in Transformer networks, the proposed method demonstrates superior performance in handling complex classification problems compared to Convolutional Neural Networks (CNNs). Experimental results show that the model achieves a recognition accuracy of up to 95.18%, exhibits high precision for each radar emitter individual, and solves the identification dilemma of similar individuals that plagues CNNs, thereby showcasing excellent recognition performance.