<i>No Radio Left Behind:</i> Radio Fingerprinting Through Deep Learning of Physical-Layer Hardware Impairments

Sankhe, Kunal; Belgiovine, Mauro; Zhou, Fan; Angioloni, Luca; Restuccia, Frank; D’Oro, Salvatore; Melodia, Tommaso; Ioannidis, Stratis; Chowdhury, Kaushik R.

doi:10.1109/tccn.2019.2949308

Cited by 150 publications

(100 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Transmitter-specific features such as phase noises, I/Q imbalance and power amplifier non-linearities have been used for transmitted identification based on WLAN data in Reference [3]. Again, such hardware-specific features have not been studied so far in the GNSS context.…”

Section: Related Work To Rffmentioning

confidence: 99%

“…In this paper, we focus the possible use of the dataset in future RF fingerprinting applications, motivated by the fact that low-cost low-power solutions for GNSS tracking and transmitter-receiver identification are more and more on demand. Radio Frequency (RF) Fingerprinting (FP) is a relatively new concept [1][2][3][4][5] focusing on identifying signals based on the hardware characteristics in the communication chain. A particular case of RFF problem is the problem of identifying transmitters for more secure communications, as a modality to distinguish genuine transmitters from 'fake' or 'attacking' ones, such as spoofers and jammers.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identifying GNSS Signals Based on Their Radio Frequency (RF) Features—A Dataset with GNSS Raw Signals Based on Roof Antennas and Spectracom Generator

Ferre

Wang

Sanz-Abia

et al. 2020

Data

View full text Add to dashboard Cite

This is a data descriptor paper for a set of raw GNSS signals collected via roof antennas and Spectracom simulator for general-purpose uses. We give one example of possible data use in the context of Radio Frequency Fingerprinting (RFF) studies for signal-type identification based on front-end hardware characteristics at transmitter or receiver side. Examples are given in this paper of achievable classification accuracy of six of the collected signal classes. The RFF is one of the state-of-the-art, promising methods to identify GNSS transmitters and receivers, and can find future applicability in anti-spoofing and anti-jamming solutions for example. The uses of the provided raw data are not limited to RFF studies, but can extend to uses such as testing GNSS acquisition and tracking, antenna array experiments, and so forth.

show abstract

Section: Related Work To Rffmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Identifying GNSS Signals Based on Their Radio Frequency (RF) Features—A Dataset with GNSS Raw Signals Based on Roof Antennas and Spectracom Generator

Ferre

Wang

Sanz-Abia

et al. 2020

Data

View full text Add to dashboard Cite

show abstract

“…We mainly select the open-source dataset used in [26]. The open-source dataset contains the wireless signals from 16 USRP X310 transmitters.…”

Section: Data Setsmentioning

confidence: 99%

“…Its robustness was proved in a wide range of SNR. Sankhe et al designed a one-dimensional convolution neural network with I/Q samples as input [26]. The network structure includes one-dimensional convolution kernels, max-pooling layers, and fully connected layers.…”

Section: Introductionmentioning

confidence: 99%

Specific Emitter Identification Against Unreliable Features Interference Based on Time-Series Classification Network Structure

Liu

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Specific emitter identification (SEI) aims to distinguish different emitter individuals based on the subtle differences in the signals. The technology can be widely applied to various wireless communication systems. Existing individual identification methods often ignore the radio frequency (RF) fingerprint information carried by the waveforms and thus can be interfered by unreliable RF features. To alleviate these issues, we devise the deep bidirectional long short-term memory (DBi-LSTM) and the one-dimensional residual convolution network with dilated convolution and squeeze-and-excitation block (Conv-OrdsNet). We exploit the combination of DBi-LSTM and Conv-OrdsNet (CoBONet) to extract temporal structure features directly from baseband in-phase and quadrature (I/Q) samples. The proposed network is able to capture the fine-grained details of signals and combine different information extracted by two networks. Moreover, we propose a data augmentation method to solve the interference of unreliable features by randomly changing the values of noise, power, frequency offset (FO), and phase offset (PO). It is worth noting that our method only needs a short slice of a steady-state signal without complex preprocessing, which reduces the cost of acquisition and calculation. Extensive experiments show that our method can effectively extract reliable RF fingerprinting features from I/Q samples and the classification results are better than most existing methods.

show abstract

“…Estimating the unintentional modulation information embedded in the transmitter [13] through the preprocessing and feature extraction of the signal is a way to amplify the subtle differences between emitters to facilitate classification in the DL framework. For steady-state signals, mainstream SEI methods focus on extracting timefrequency domain features, mapping the time-domain waveform to the time-frequency plane, and analyzing the time-frequency joint information.…”

Section: Introductionmentioning

confidence: 99%

Specific Emitter Identification Based on Software-Defined Radio and Decision Fusion

Tan¹,

Yan²,

Zhang³

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Specific emitter identification (SEI) uses the unintentional modulation information carried by the emitter waveform, i.e., radio frequency fingerprints, to realize the matching identification of the received signal and its corresponding emitter. We propose an emitter identification scheme based on deep learning (DL). The received signal is subjected to time-varying filtered empirical mode decomposition (tvf-EMD). The obtained intrinsic mode functions (IMFs) are subjected to Hilbert transformation to obtain a 3D-Hilbert spectrum, using amplitude frequency aggregation characteristics obtained by projection to express the nonstationary information of the signal and calculate its bispectral diagonal slice as the second feature. A squeeze-and-excitation block (SEB) is introduced to a convolutional neural network (CNN) to focus on the effective area in the feature map, and support vector machine (SVM) is used to fuse the decision information of the two types of features. We collect four types of steady-state signals on a software-defined radio (SDR) sensor platform based on GNU Radio and universal software radio peripherals (USRP), and study the algorithm performance. The results show that our scheme has strong generalization, high recognition accuracy, and broad application prospects compared with other SEI methods.INDEX TERMS Specific emitter identification, empirical mode decomposition (EMD), bispectrum, software-defined radio, channel attention, decision fusion. Preprocess Feature Extraction Receiver DL classifier Signals From Multiple Emitters FIGURE 1. Non-cooperative SEI system based on deep learning.

show abstract

No Radio Left Behind: Radio Fingerprinting Through Deep Learning of Physical-Layer Hardware Impairments

Cited by 150 publications

References 25 publications

Identifying GNSS Signals Based on Their Radio Frequency (RF) Features—A Dataset with GNSS Raw Signals Based on Roof Antennas and Spectracom Generator

Identifying GNSS Signals Based on Their Radio Frequency (RF) Features—A Dataset with GNSS Raw Signals Based on Roof Antennas and Spectracom Generator

Specific Emitter Identification Against Unreliable Features Interference Based on Time-Series Classification Network Structure

Specific Emitter Identification Based on Software-Defined Radio and Decision Fusion

Contact Info

Product

Resources

About