Specific Emitter Identification for Communications Transmitter Using Multi-measurements

Huang, Guangquan; Yuan, Yingjun; Xiang, Wu; Huang, Zhitao

doi:10.1007/s11277-016-3696-8

Cited by 16 publications

(8 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So it is usually used for mining useful information of signals. Common entropy feature mining methods include methods based on permutation entropy [40], multi-dimensional permutation entropy [41], and normalized permutation entropy vector (NPEV) [42]. The author in [36] uses characteristics caused by hardware defects to distinguish devices, but did not deeply study the impact of different components imperfections on RFF.…”

Section: B Permutation Entropy Methodsmentioning

confidence: 99%

Wireless Transmitter Identification Based on Device Imperfections

et al. 2020

View full text Add to dashboard Cite

Mining useful patterns from databases is an important research topic. The research in utility mining mostly focuses on discovering patterns of high value in large databases, and analyzing the important factors in a data mining process. This idea is applied to the wireless device identification in this paper. Radio Frequency Fingerprint (RFF) reflects differences between transmitter hardware components. It contains rich non-linear characteristics of the internal components of the transmitter. Small differences and inaccuracies in the manufacturing process determine the unique characteristic contained in the transmitted signal. The device can be identified by the signal transmitted by the wireless device. In this paper, the generation mechanism of RFF is analyzed and two pattern mining algorithms are used to extract useful information from wireless signals for device identification. Then, a real communication transmitter link is established to study the effect of different components of a transmitter. The signals are acquired from the transmitters with different components replaced, including the amplifier, the bandpass filter, and the local oscillator. Finally, the influence of different components and pattern mining methods are evaluated.

show abstract

Section: B Permutation Entropy Methodsmentioning

confidence: 99%

Wireless Transmitter Identification Based on Device Imperfections

et al. 2020

View full text Add to dashboard Cite

show abstract

“…1) These methods require efforts to manually extract features or high-order statistics, the quality of handcraft features dominates device identification performances. E.g., authors in [124] show that the combination of permutation entropy [125] and K-NN even surpasses combination of bispectrum [126] and SVM in [111]. 2) Experiments are conducted in rational environments with a limited number (less than 30) of IoT devices.…”

Section: B Feature-based Statistical Learning For Specific Device Ide...mentioning

confidence: 99%

Machine Learning for the Detection and Identification of Internet of Things (IoT) Devices: A Survey

Liu

Wang

et al. 2021

Preprint

View full text Add to dashboard Cite

The Internet of Things (IoT) is becoming an indispensable part of everyday life, enabling a variety of emerging services and applications. However, the presence of rogue IoT devices has exposed the IoT to untold risks with severe consequences. The first step in securing the IoT is detecting rogue IoT devices and identifying legitimate ones. Conventional approaches use cryptographic mechanisms to authenticate and verify legitimate devices' identities. However, cryptographic protocols are not available in many systems. Meanwhile, these methods are less effective when legitimate devices can be exploited or encryption keys are disclosed. Therefore, non-cryptographic IoT device identification and rogue device detection become efficient solutions to secure existing systems and will provide additional protection to systems with cryptographic protocols. Noncryptographic approaches require more effort and are not yet adequately investigated. In this paper, we provide a comprehensive survey on machine learning technologies for the identification of IoT devices along with the detection of compromised or falsified ones from the viewpoint of passive surveillance agents or network operators. We classify the IoT device identification and detection into four categories: device-specific pattern recognition, Deep Learning enabled device identification, unsupervised device identification, and abnormal device detection. Meanwhile, we discuss various ML-related enabling technologies for this purpose. These enabling technologies include learning algorithms, feature engineering on network traffic traces and wireless signals, continual learning, and abnormality detection.

show abstract

“…In Refs. [17,18], the normalized permutation entropy (NPE) was used as rf fingerprints to identify the unique emitter of given signals. The NPE-based method also needs much data to obtain a reliable statistic and the embedding dimension would affect the classification performance.…”

Section: Pacsmentioning

confidence: 99%

Specific Emitter Identification Based on Visibility Graph Entropy

Zhu¹,

Gan²

2018

Chinese Phys. Lett.

View full text Add to dashboard Cite

The specific emitter identification (SEI) technique determines the unique emitter of a given signal by using some external feature measurements of the signal. It has recently attracted a great deal of attention because many applications can benefit from it. This work addresses the SEI problem using two methods, namely, the normalized visibility graph entropy (NVGE) and the normalized horizontal visibility graph entropy (NHVGE) based on treating emitters as nonlinear dynamical systems. Firstly, the visibility graph (VG) and the horizontal visibility graph (HVG) are used to convert the instantaneous amplitude, phase and frequency of received signals into graphs. Then, based on the information captured by the VG and the HVG, the normalized Shannon entropy (NSE) calculated from the corresponding degree distributions are utilized as the rf fingerprint. Finally, four emitters from the same manufacturer are utilized to evaluate the performance of the two methods. Experimental results demonstrate that both the NHVGE-based method and NVGE-based method are quite effective and they perform much better than the method based on the normalized permutation entropy (NPE) in the case of a small amount of data. The NVGE-based method performs better than the NHVGE-based method since the VG can extract more information than the HVG does. Moreover, our methods do not distinguish between the transient signal and the steady-state signal, making it practical.

show abstract

Specific Emitter Identification for Communications Transmitter Using Multi-measurements

Cited by 16 publications

References 22 publications

Wireless Transmitter Identification Based on Device Imperfections

Wireless Transmitter Identification Based on Device Imperfections

Machine Learning for the Detection and Identification of Internet of Things (IoT) Devices: A Survey

Specific Emitter Identification Based on Visibility Graph Entropy

Contact Info

Product

Resources

About