Transfer Learning for Automatic Modulation Recognition Using a Few Modulated Signal Samples

Wensheng Lin; Dongbin Hou; Junsheng Huang; Lixin Li; Zhu Han

doi:10.1109/TVT.2023.3267270

Transfer Learning for Automatic Modulation Recognition Using a Few Modulated Signal Samples

Wensheng Lin
, Dongbin Hou
, Junsheng Huang
, Lixin Li
, Zhu Han

School of Electronics and Information

Northwestern Polytechnical University Xian
University of Houston
Kyung Hee University

Research output: Contribution to journal › Article › peer-review

58 Scopus citations

Abstract

This letter proposes a transfer learning model for automatic modulation recognition (AMR) with only a few modulated signal samples. The transfer model is trained with the audio signal UrbanSound8K as the source domain, and then fine-tuned with a few modulated signal samples as the target domain. For improving the classification performance, the signal-to-noise ratio (SNR) is utilized as a feature to facilitate the classification of signals. Simulation results indicate that the transfer model has a significant superiority in terms of classification accuracy.

Original language	English
Pages (from-to)	12391-12395
Number of pages	5
Journal	IEEE Transactions on Vehicular Technology
Volume	72
Issue number	9
DOIs	https://doi.org/10.1109/TVT.2023.3267270
State	Published - 1 Sep 2023

Keywords

automatic modulation recognition
convolutional neural network
deep learning
few-shot learning
Transfer learning

Access to Document

10.1109/TVT.2023.3267270

Cite this

@article{5afcb3cebb6b4c05af2a5d8001dc7cb2,

title = "Transfer Learning for Automatic Modulation Recognition Using a Few Modulated Signal Samples",

abstract = "This letter proposes a transfer learning model for automatic modulation recognition (AMR) with only a few modulated signal samples. The transfer model is trained with the audio signal UrbanSound8K as the source domain, and then fine-tuned with a few modulated signal samples as the target domain. For improving the classification performance, the signal-to-noise ratio (SNR) is utilized as a feature to facilitate the classification of signals. Simulation results indicate that the transfer model has a significant superiority in terms of classification accuracy.",

keywords = "automatic modulation recognition, convolutional neural network, deep learning, few-shot learning, Transfer learning",

author = "Wensheng Lin and Dongbin Hou and Junsheng Huang and Lixin Li and Zhu Han",

note = "Publisher Copyright: {\textcopyright} 1967-2012 IEEE.",

year = "2023",

month = sep,

day = "1",

doi = "10.1109/TVT.2023.3267270",

language = "英语",

volume = "72",

pages = "12391--12395",

journal = "IEEE Transactions on Vehicular Technology",

issn = "0018-9545",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "9",

}

TY - JOUR

T1 - Transfer Learning for Automatic Modulation Recognition Using a Few Modulated Signal Samples

AU - Lin, Wensheng

AU - Hou, Dongbin

AU - Huang, Junsheng

AU - Li, Lixin

AU - Han, Zhu

PY - 2023/9/1

Y1 - 2023/9/1

N2 - This letter proposes a transfer learning model for automatic modulation recognition (AMR) with only a few modulated signal samples. The transfer model is trained with the audio signal UrbanSound8K as the source domain, and then fine-tuned with a few modulated signal samples as the target domain. For improving the classification performance, the signal-to-noise ratio (SNR) is utilized as a feature to facilitate the classification of signals. Simulation results indicate that the transfer model has a significant superiority in terms of classification accuracy.

AB - This letter proposes a transfer learning model for automatic modulation recognition (AMR) with only a few modulated signal samples. The transfer model is trained with the audio signal UrbanSound8K as the source domain, and then fine-tuned with a few modulated signal samples as the target domain. For improving the classification performance, the signal-to-noise ratio (SNR) is utilized as a feature to facilitate the classification of signals. Simulation results indicate that the transfer model has a significant superiority in terms of classification accuracy.

KW - automatic modulation recognition

KW - convolutional neural network

KW - deep learning

KW - few-shot learning

KW - Transfer learning

UR - https://www.scopus.com/pages/publications/85153347026

U2 - 10.1109/TVT.2023.3267270

DO - 10.1109/TVT.2023.3267270

M3 - 文章

AN - SCOPUS:85153347026

SN - 0018-9545

VL - 72

SP - 12391

EP - 12395

JO - IEEE Transactions on Vehicular Technology

JF - IEEE Transactions on Vehicular Technology

IS - 9

ER -

Transfer Learning for Automatic Modulation Recognition Using a Few Modulated Signal Samples

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this