Few-Shot Specific Emitter Identification via Integrated Complex Variational Mode Decomposition and Spatial Attention Transfer

TL;DR

The paper tackles few-shot specific emitter identification (FS-SEI) by introducing an integrated framework that combines signal reconstruction through Integrated Complex Variational Mode Decomposition (ICVMD), a temporal convolutional backbone (CNN-TCN), and a parallel spatial attention mechanism with attention transfer (SAT). This approach enables robust feature extraction from hardware-based RF fingerprints with minimal labeled data, achieving 96% accuracy using only 10 symbols without prior information on public data. Ablation studies on simulated data and validation on a real ZigBee dataset demonstrate the method's superiority and transferability, highlighting the benefits of complex signal decomposition, sequential modeling, and cross-task attention transfer. Overall, ICVMD-SAT reduces reliance on large labeled datasets and prior knowledge, offering a practical, transfer-friendly solution for real-world SEI scenarios.

Abstract

Specific emitter identification (SEI) utilizes passive hardware characteristics to authenticate transmitters, providing a robust physical-layer security solution. However, most deep-learning-based methods rely on extensive data or require prior information, which poses challenges in real-world scenarios with limited labeled data. We propose an integrated complex variational mode decomposition algorithm that decomposes and reconstructs complex-valued signals to approximate the original transmitted signals, thereby enabling more accurate feature extraction. We further utilize a temporal convolutional network to effectively model the sequential signal characteristics, and introduce a spatial attention mechanism to adaptively weight informative signal segments, significantly enhancing identification performance. Additionally, the branch network allows leveraging pre-trained weights from other data while reducing the need for auxiliary datasets. Ablation experiments on the simulated data demonstrate the effectiveness of each component of the model. An accuracy comparison on a public dataset reveals that our method achieves 96% accuracy using only 10 symbols without requiring any prior knowledge.

Figures (12)

• Figure 1: The signal generation and transmission process.
• Figure 2: The structure of the ICVMD-SAT method.
• Figure 3: Causal dilation convolution structure ($d={1,2,4,8}$, $l=2$).
• Figure 4: The structure of a simple TCN block.
• Figure 5: A schematic diagram of the spatial attention mechanism.