Machine Learning for RF Fingerprinting
Extraction and Identification
of Soft-Defined Radio Devices
Su Hu
1(B) , Pei Wang
1 , Yaping Peng
1 , Di Lin
1 , Yuan Gao
2 , Jiang Cao
2 ,
and Bin Yu
3
1 University of Electronic Science and Technology of China,
Chengdu 611731, Sichuan, China
husu@uestc.edu.cn
2 Academy of Military Science of PLA, Beijing 100090, China
3 Beijing Samsung Telecom R&D Center, Beijing 100081, China
Abstract. Radio frequency (RF) fingerprinting technology has been
developed as a unique method for maintaining security based on physical
layer characteristics. In this paper, we propose the RF fingerprinting by
extracting the parameter characteristics such as information dimension,
constellation feature, and phase noise spectrum in the transmitted information when applied to the universal software radio peripheral (USRP)
software-defined radio (SDR) platform. To achieve a great performance
improvement of classification, not only the traditional support vector
machine (SVM) classifier, but also the machine-based integrated classifier bagged tree and the adaptive weighting algorithm weighted k-nearest
neighbor (KNN) are both discussed. It is demonstrated that the proposed
method achieves good classification performance under different signalto-noise ratios (SNR).
Keywords: RF fingerprinting · Wireless communication ·
Software-defined radio · Characteristic parameter · Classifier
1 Introduction
With the continuous popularization of mobile communication devices and the
rapid development of the Internet of things (IoT) technology, wireless communication plays an irreplaceable role in both military and civilian applications.
However, due to its openness, wireless networks are more vulnerable to largescale malicious attacks than traditional wired networks. Meanwhile owing to
artificial intelligence technology [1], several new low-cost devices are sensitive
to computational complexity. And traditional methods by employ IP or MAC
addresses as identity are not effective as usual. Therefore, in order to reduce
potential threats from malicious users, it is urgent to find a new type of security
mechanism to identify authorized users and unauthorized users effectively.
c
Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 189–204, 2020.
https://doi.org/10.1007/978-981-15-0187-6_22
Extraction and Identification
of Soft-Defined Radio Devices
Su Hu
1(B) , Pei Wang
1 , Yaping Peng
1 , Di Lin
1 , Yuan Gao
2 , Jiang Cao
2 ,
and Bin Yu
3
1 University of Electronic Science and Technology of China,
Chengdu 611731, Sichuan, China
husu@uestc.edu.cn
2 Academy of Military Science of PLA, Beijing 100090, China
3 Beijing Samsung Telecom R&D Center, Beijing 100081, China
Abstract. Radio frequency (RF) fingerprinting technology has been
developed as a unique method for maintaining security based on physical
layer characteristics. In this paper, we propose the RF fingerprinting by
extracting the parameter characteristics such as information dimension,
constellation feature, and phase noise spectrum in the transmitted information when applied to the universal software radio peripheral (USRP)
software-defined radio (SDR) platform. To achieve a great performance
improvement of classification, not only the traditional support vector
machine (SVM) classifier, but also the machine-based integrated classifier bagged tree and the adaptive weighting algorithm weighted k-nearest
neighbor (KNN) are both discussed. It is demonstrated that the proposed
method achieves good classification performance under different signalto-noise ratios (SNR).
Keywords: RF fingerprinting · Wireless communication ·
Software-defined radio · Characteristic parameter · Classifier
1 Introduction
With the continuous popularization of mobile communication devices and the
rapid development of the Internet of things (IoT) technology, wireless communication plays an irreplaceable role in both military and civilian applications.
However, due to its openness, wireless networks are more vulnerable to largescale malicious attacks than traditional wired networks. Meanwhile owing to
artificial intelligence technology [1], several new low-cost devices are sensitive
to computational complexity. And traditional methods by employ IP or MAC
addresses as identity are not effective as usual. Therefore, in order to reduce
potential threats from malicious users, it is urgent to find a new type of security
mechanism to identify authorized users and unauthorized users effectively.
c
Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 189–204, 2020.
https://doi.org/10.1007/978-981-15-0187-6_22