A Radio Frequency Fingerprint Identification Method for Wireless Devices Based on ShuffleNet-SVM

Radio frequency fingerprint identification can be used for device authentication, network access management, malicious device tracing, and identity spoofing attacks. The feature representation and pattern recognition capabilities of deep learning algorithms are of great significance for improving the identification capability of radio frequency fingerprint-based wireless devices. Aiming at the problem of radio frequency fingerprint identification of wireless devices, this paper proposes a radio frequency fingerprint identification method for wireless devices based on ShuffleNet-SVM. Firstly, the original signal of the wireless device is processed by the constant Q transform. Secondly, the obtained CQT spectrum is input into the ShuffleNet model for training to achieve adaptive feature extraction, and then the output of the convolution layer is extracted as the RF fingerprint feature vector, which is used as the input of the support vector machine. Finally, the output of the identification result can realize the radio frequency fingerprint identification of the wireless device. Based on the Bluetooth device data set, a series of experiments verify that the proposed method is superior to the comparison models and other time-frequency processing methods in terms of accuracy and robustness.