BiGRU-Attention-Based Specific Emitter Identification Method

Specific emitter identification plays a pivotal role in electronic countermeasures within the realm of complex electromagnetic environments. This study introduces a novel method for specific emitter identification, leveraging the capabilities of BiGRU-Attention (BiGRU-Att). The process begins with the segmentation of steady-state signal slices from the targeted emitter. Subsequently, six distinct features, including wavelet entropy and amplitude skewness within the timefrequency domain, are meticulously extracted. These extracted sequence feature data are then employed to train the model, culminating in the acquisition of the identification outcome. The methodology was rigorously tested using TETRA communication signals from nine distinct emitters. Experimental findings affirm that the model proficiently harnesses the timing information inherent in the specific emitter's signal, substantially enhancing the identification accuracy. This approach not only underscores the effectiveness of integrating BiGRU and attention mechanisms for signal classification but also sets a precedent for future research in specific emitter identification under intricate electromagnetic conditions.