This paper presents a physical layer wireless device identification method that uses a convolutional neural network (CNN) operating on a quadrant IQ transition image. This work introduces classification and detection tasks in one process. The proposed method can identify IoT wireless devices by exploiting their RF fingerprints, a technology to identify wireless devices by using unique variations in analog signals. We propose a quadrant IQ image technique to reduce the size of CNN while maintaining accuracy. The CNN utilizes the IQ transition image, which image processing cut out into four-part. An over-the-air experiment is performed on six Zigbee wireless devices to confirm the proposed identification method's validity. The measurement results demonstrate that the proposed method can achieve 99% accuracy with the light-weight CNN model with 36,500 weight parameters in serial use and 146,000 in parallel use. Furthermore, the proposed threshold algorithm can verify the authenticity using one classifier and achieved 80% accuracy for further secured wireless communication. This work also introduces the identification of expanded signals with SNR between 10 to 30dB. As a result, at SNR values above 20dB, the proposals achieve classification and detection accuracies of 87% and 80%, respectively.
Hiro TAMURA
Tokyo Institute of Technology
Kiyoshi YANAGISAWA
Tokyo Institute of Technology
Atsushi SHIRANE
Tokyo Institute of Technology
Kenichi OKADA
Tokyo Institute of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Hiro TAMURA, Kiyoshi YANAGISAWA, Atsushi SHIRANE, Kenichi OKADA, "Performance Evaluation of Classification and Verification with Quadrant IQ Transition Image" in IEICE TRANSACTIONS on Communications,
vol. E105-B, no. 5, pp. 580-587, May 2022, doi: 10.1587/transcom.2021EBP3087.
Abstract: This paper presents a physical layer wireless device identification method that uses a convolutional neural network (CNN) operating on a quadrant IQ transition image. This work introduces classification and detection tasks in one process. The proposed method can identify IoT wireless devices by exploiting their RF fingerprints, a technology to identify wireless devices by using unique variations in analog signals. We propose a quadrant IQ image technique to reduce the size of CNN while maintaining accuracy. The CNN utilizes the IQ transition image, which image processing cut out into four-part. An over-the-air experiment is performed on six Zigbee wireless devices to confirm the proposed identification method's validity. The measurement results demonstrate that the proposed method can achieve 99% accuracy with the light-weight CNN model with 36,500 weight parameters in serial use and 146,000 in parallel use. Furthermore, the proposed threshold algorithm can verify the authenticity using one classifier and achieved 80% accuracy for further secured wireless communication. This work also introduces the identification of expanded signals with SNR between 10 to 30dB. As a result, at SNR values above 20dB, the proposals achieve classification and detection accuracies of 87% and 80%, respectively.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2021EBP3087/_p
Copy
@ARTICLE{e105-b_5_580,
author={Hiro TAMURA, Kiyoshi YANAGISAWA, Atsushi SHIRANE, Kenichi OKADA, },
journal={IEICE TRANSACTIONS on Communications},
title={Performance Evaluation of Classification and Verification with Quadrant IQ Transition Image},
year={2022},
volume={E105-B},
number={5},
pages={580-587},
abstract={This paper presents a physical layer wireless device identification method that uses a convolutional neural network (CNN) operating on a quadrant IQ transition image. This work introduces classification and detection tasks in one process. The proposed method can identify IoT wireless devices by exploiting their RF fingerprints, a technology to identify wireless devices by using unique variations in analog signals. We propose a quadrant IQ image technique to reduce the size of CNN while maintaining accuracy. The CNN utilizes the IQ transition image, which image processing cut out into four-part. An over-the-air experiment is performed on six Zigbee wireless devices to confirm the proposed identification method's validity. The measurement results demonstrate that the proposed method can achieve 99% accuracy with the light-weight CNN model with 36,500 weight parameters in serial use and 146,000 in parallel use. Furthermore, the proposed threshold algorithm can verify the authenticity using one classifier and achieved 80% accuracy for further secured wireless communication. This work also introduces the identification of expanded signals with SNR between 10 to 30dB. As a result, at SNR values above 20dB, the proposals achieve classification and detection accuracies of 87% and 80%, respectively.},
keywords={},
doi={10.1587/transcom.2021EBP3087},
ISSN={1745-1345},
month={May},}
Copy
TY - JOUR
TI - Performance Evaluation of Classification and Verification with Quadrant IQ Transition Image
T2 - IEICE TRANSACTIONS on Communications
SP - 580
EP - 587
AU - Hiro TAMURA
AU - Kiyoshi YANAGISAWA
AU - Atsushi SHIRANE
AU - Kenichi OKADA
PY - 2022
DO - 10.1587/transcom.2021EBP3087
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E105-B
IS - 5
JA - IEICE TRANSACTIONS on Communications
Y1 - May 2022
AB - This paper presents a physical layer wireless device identification method that uses a convolutional neural network (CNN) operating on a quadrant IQ transition image. This work introduces classification and detection tasks in one process. The proposed method can identify IoT wireless devices by exploiting their RF fingerprints, a technology to identify wireless devices by using unique variations in analog signals. We propose a quadrant IQ image technique to reduce the size of CNN while maintaining accuracy. The CNN utilizes the IQ transition image, which image processing cut out into four-part. An over-the-air experiment is performed on six Zigbee wireless devices to confirm the proposed identification method's validity. The measurement results demonstrate that the proposed method can achieve 99% accuracy with the light-weight CNN model with 36,500 weight parameters in serial use and 146,000 in parallel use. Furthermore, the proposed threshold algorithm can verify the authenticity using one classifier and achieved 80% accuracy for further secured wireless communication. This work also introduces the identification of expanded signals with SNR between 10 to 30dB. As a result, at SNR values above 20dB, the proposals achieve classification and detection accuracies of 87% and 80%, respectively.
ER -