This project builds a length-versatile and noise-robust LoRa radio frequency fingerprint identification (RFFI) system. The LoRa signals are collected from 10 commercial-off-the-shelf LoRa devices, with the spreading factor (SF) set to 7, 8, 9, respectively. The packet preamble part and device labels are provided.
If the part of the dataset/codes contributes to your project, please cite:
[1] G. Shen, J. Zhang, A. Marshall, M. Valkama, and J. Cavallaro. “Towards Length-Versatile and Noise-Robust Radio Frequency Fingerprint Identification,” IEEE Trans. Inf. Forensics Security, 2023.
@article{shen2023length,
title={Towards Length-Versatile and Noise-Robust Radio Frequency Fingerprint Identification},
author={Shen, Guanxiong and Zhang, Junqing and Marshall, Alan and Valkama, Mikko and Cavallaro, Joseph},
journal={IEEE Trans. Inf. Forensics Security},
year={2023}
}
Transmitters: device 31-35 LoPy4, device 36-40 Dragino LoRa shield.
Receiver: USRP N210 software-defined radio (SDR).
| Name | Number of Packets Per Device | Spreading Factor |
|---|---|---|
| sf_7_train.h5 | 2,500 | 7 |
| sf_8_train.h5 | 2,500 | 8 |
| sf_9_train.h5 | 2,500 | 9 |
| sf_7_test.h5 | 500 | 7 |
| sf_8_test.h5 | 500 | 8 |
| sf_9_test.h5 | 500 | 9 |
a) Install Required Packages
Please find the 'requirement.txt' file to install the required packages.
b) Download Dataset
Please downlaod the dataset and put it in the project folder. The download link is https://ieee-dataport.org/documents/lorarffidatasetdifferentspreadingfactors.
c) Operating System
This project is built entirely on the Windows operating system. There may be unexpected issues on other operating systems.
The function 'train()' can train a length-versatile neural network, i.e., LSTM, GRU, Transformer or 'Flatten-free CNN'. Please change the variable 'model_type' to specify the type of the trained neural network.
The function 'inference()' can evaluate the trained neural network. It returns the overall accuracy and a confusion matrix. Please change the variable 'snr_awgn' to specify the range of artificial noise added to the test data.
The dataset and code is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Please contact the following email addresses if you have any questions:
Guanxiong.Shen AT liverpool.ac.uk
Junqing.Zhang AT liverpool.ac.uk