Radio Frequency Fingerprint Identification for Security in Low-Cost IoT Devices

@article{Shen2021RadioFF,
  title={Radio Frequency Fingerprint Identification for Security in Low-Cost IoT Devices},
  author={Guanxiong Shen and Junqing Zhang and Alan J. Marshall and Mikko Valkama and Joe Cavallaro},
  journal={2021 55th Asilomar Conference on Signals, Systems, and Computers},
  year={2021},
  pages={309-313}
}
Radio frequency fingerprint identification (RFFI) can uniquely classify wireless devices by analyzing the received signal distortions caused by the intrinsic hardware impairments. The state-of-the-art deep learning techniques such as convolutional neural network (CNN) have been adopted to classify IoT devices with high accuracy. However, deep learning-based RFFI requires input data of a fixed size. In addition, many IoT devices work in low signal-to-noise ratio (SNR) scenarios but the low SNR… 

Figures from this paper

Towards Length-Versatile and Noise-Robust Radio Frequency Fingerprint Identification
TLDR
The reason why MLP- and CNN-based RFFI systems are constrained by the input size is analyzed and four neural networks that can process signals of variable lengths are proposed, namely flatten-free CNN, long short-term memory (LSTM) network, gated recurrent unit (GRU) network and transformer.
LoRaWAN Physical Layer-Based Attacks and Countermeasures, A Review
TLDR
This work aims to provide a comprehensive review of various vulnerabilities, countermeasures and security enhancing features concerning the LoRaWAN physical layer and identifies research gaps as well as promising future research directions.
Towards Receiver-Agnostic and Collaborative Radio Frequency Fingerprint Identification
TLDR
The results show that receiver-agnostic training enables the trained neural network to become robust to changes in receiver characteristics, and the collaborative inference improves classification accuracy by up to 20% beyond a single-receiver RFFI system and can bring a 40% improvement for under-performing receivers.

References

SHOWING 1-10 OF 18 REFERENCES
Radio Frequency Fingerprint Identification for LoRa Using Deep Learning
TLDR
A hybrid classifier that can adjust the prediction of deep learning models with the estimated CFO is designed to further increase the classification accuracy of the deep learning-based RFFI scheme for Long Range (LoRa) systems.
A Robust RF Fingerprinting Approach Using Multisampling Convolutional Neural Network
TLDR
A multisampling convolutional neural network (MSCNN) to extract RF fingerprint from the selected ROI for classifying ZigBee devices and is robust over a wide range of SNRs under the LOS scenarios as well as under the NLOS scenarios.
Towards Scalable and Channel-Robust Radio Frequency Fingerprint Identification for LoRa
TLDR
This paper proposes a scalable and channel-robust RFFI framework achieved by deep learning powered radio frequency fingerprint extractor and channel independent features and leverages deep metric learning to train an RFF extractor, which has excellent generalization ability and can extract RFFs from previously unseen devices.
Enhanced RF Fingerprinting for IoT Devices with Recurrent Neural Networks
TLDR
This paper develops a physical-layer authentication technique using a neural network structure with both convolutional and recurrent components to distinguish transmissions originating from a particular target device from all others and demonstrates strong performance in a realistic multipath channel environment.
Exposing the Fingerprint: Dissecting the Impact of the Wireless Channel on Radio Fingerprinting
TLDR
Experimental results conclude that the wireless channel impacts the classification accuracy significantly, and equalizing I/Q data can increase the accuracy to a significant extent when the number of devices increases significantly.
On Radio Frequency Fingerprint Identification for DSSS Systems in Low SNR Scenarios
TLDR
In the scheme, a signal preprocessing method, information data estimation-based stacking algorithm, is proposed, which leverages the repeated spreading sequences and stacks them together to eliminate the noise and interference effect.
DeepLoRa: Fingerprinting LoRa Devices at Scale Through Deep Learning and Data Augmentation
TLDR
This work collects and publicly shared more than 1TB of waveform data from 100 bit-similar devices over different deployment scenarios (outdoor vs. indoor) and spanning several days, and proposes a novel data augmentation technique called DeepLoRa to enhance the LoRa RFFP performance.
Open Set Wireless Transmitter Authorization: Deep Learning Approaches and Dataset Considerations
TLDR
This article forms the problem of recognizing authorized transmitters and rejecting new transmitters as open set recognition and anomaly detection, and considers approaches based on one and several binary classifiers, multiclass classifier, and signal reconstruction.
RF-Based Low-SNR Classification of UAVs Using Convolutional Neural Networks
TLDR
A single model using spectrogram images of noisy signals and tuning the CNN model parameters achieves a classification accuracy varying from 92% to 100% for an SNR range from -10 dB to 30 dB, which significantly outperforms the existing approaches to the best knowledge.
More Is Better: Data Augmentation for Channel-Resilient RF Fingerprinting
TLDR
This article proposes a data augmentation step within the training pipeline that exposes the DNN to many simulated channel and noise variations that are not present in the original dataset.
...
...