Research Article
A RF Fingerprint Clustering Method Based on Automatic Feature Extractor
@INPROCEEDINGS{10.1007/978-3-031-30237-4_10, author={Pinhong Xiao and Di Lin and Mengjuan Wang}, title={A RF Fingerprint Clustering Method Based on Automatic Feature Extractor}, proceedings={Machine Learning and Intelligent Communication. 7th EAI International Conference, MLICOM 2022, Virtual Event, October 23-24, 2022, Proceedings}, proceedings_a={MLICOM}, year={2023}, month={4}, keywords={RF fingerprint identification Semi-supervised Learning Feature extraction}, doi={10.1007/978-3-031-30237-4_10} }- Pinhong Xiao
Di Lin
Mengjuan Wang
Year: 2023
A RF Fingerprint Clustering Method Based on Automatic Feature Extractor
MLICOM
Springer
DOI: 10.1007/978-3-031-30237-4_10
Abstract
RF fingerprint technology has received extensive attention and research in recent years due to its immutable nature. RF fingerprinting technology can be used as a wireless network security mechanism alone or combined with existing security mechanisms to enhance wireless network security. The early RF fingerprint research widely used the method of artificial feature extraction, but this method relies too much on expert experience. This paper proposes a semi-supervised learning approach for RF fingerprint recognition. This work directly uses the original I/Q sequence data, designs a fingerprint extractor based on the convolutional neural network (CNN), and uses K-means and DBSCAN algorithms to cluster the fingerprints. The experimental results demonstrate that after training with a small amount of labeled data, the fingerprint extractor can effectively extract features of unknown signals, and these features can well allow unknown similar devices to be clustered together by the clustering algorithm.
