Radio frequency fingerprint identification (RFFI) is an emerging device authentication technique that relies on the intrinsic hardware characteristics of wireless devices. This paper designs a deep learning-based RFFI scheme for Long Range (LoRa) systems. Firstly, the instantaneous carrier frequency offset (CFO) is found to drift, which could result in misclassification and significantly compromise the stability of the deep learning-based RFFI system. CFO compensation is demonstrated to be effective mitigation. Secondly, three signal representations for deep learning-based RFFI are investigated in time, frequency, and time-frequency domains, namely in-phase and quadrature (IQ) samples, fast Fourier transform (FFT) results and spectrograms, respectively. For these signal representations, three deep learning models are implemented, i.e., multilayer perceptron (MLP), long short-term memory (LSTM) network and convolutional neural network (CNN), in order to explore an optimal framework. Finally, a hybrid classifier that can adjust the prediction of deep learning models with the estimated CFO is designed to further increase the classification accuracy. The CFO will not change dramatically over several continuous days, hence it can be used to correct predictions when the estimated CFO is much different from the reference one. Experimental evaluation is performed in real wireless environments involving 25 LoRa devices and a Universal Software Radio Peripheral (USRP) N210 platform. The spectrogram-CNN model is found to be optimal for classifying LoRa devices which can reach an accuracy of 96.40% with the least complexity and training time.

中文翻译：

---

使用深度学习的 LoRa 射频指纹识别


射频指纹识别（RFFI）是一种新兴的设备身份验证技术，依赖于无线设备固有的硬件特性。本文为远程（LoRa）系统设计了一种基于深度学习的 RFFI 方案。首先，瞬时载波频率偏移（CFO）会发生漂移，这可能会导致错误分类并严重损害基于深度学习的 RFFI 系统的稳定性。事实证明，首席财务官薪酬是有效的缓解措施。其次，在时域、频域和时频域中研究了基于深度学习的 RFFI 的三种信号表示，分别是同相和正交 (IQ) 样本、快速傅立叶变换 (FFT) 结果和频谱图。对于这些信号表示，实现了三种深度学习模型，即多层感知器（MLP）、长短期记忆（LSTM）网络和卷积神经网络（CNN），以探索最佳框架。最后，设计了一种混合分类器，可以根据估计的 CFO 调整深度学习模型的预测，以进一步提高分类精度。 CFO 在连续几天内不会发生显着变化，因此当估计的 CFO 与参考值相差很大时，它可以用于纠正预测。实验评估是在涉及 25 个 LoRa 设备和通用软件无线电外设 (USRP) N210 平台的真实无线环境中进行的。频谱图-CNN 模型被发现是 LoRa 设备分类的最佳模型，其准确率可以达到 96.40%，且复杂度和训练时间最少。