Radio Frequency Emitter Identification Based on Ensemble Learning
Qi Cheng, Jie Li, Xiaoli Gao, Quan Zhi Zhou, Heng Liu, Yang Liu · 2023
In order to accurately identify and authenticate radio frequency (RF) emitters, preventing the occurrence of problems such as device cloning, replay attacks and user identity impersonation, an identification method based on time- frequency feature extraction and ensemble learning is proposed. First, the short-time Fourier transform (STFT) is applied on the time-domain signals to generate the time-frequency images which carry the essential characteristics of the signals. Then three sub-models of Resnet34, Inceptionv3 and Densenet121 are trained on the data set with the features extracted by each sub- model and fused to obtain a comprehensive model. The experimental results show that the overall recognition accuracy of the proposed algorithm for 10 LoRa RF emitters outperforms the sub-models applied individually.