The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Pengecaman modulasi automatik (AMR) memainkan peranan penting dalam sistem komunikasi moden. Disebabkan oleh kemajuan teknik pembelajaran mendalam (DL) baru-baru ini, aplikasi DL telah dikaji secara meluas dalam AMR, dan sejumlah besar algoritma DL-AMR dengan kadar pengecaman yang tinggi telah dibangunkan. Kebanyakan model algoritma DL-AMR mempunyai ketepatan pengecaman yang tinggi tetapi mempunyai banyak parameter dan merupakan model yang besar dan kompleks, yang menjadikannya sukar untuk digunakan pada platform yang dikekang sumber, seperti platform satelit. Beberapa model algoritma DL-AMR yang ringan dan kerumitan rendah juga berjuang untuk memenuhi keperluan ketepatan. Berdasarkan ini, kertas kerja ini mencadangkan model algoritma DL-AMR yang ringan dan kadar pengecaman tinggi yang dipanggil Rangkaian Memori Jangka Pendek Jangka Pendek (LDLSTM) Lightweight Densely Connected Convolutional Network (DenseNet). Lata model DenseNet dan LSTM boleh mencapai ketepatan pengecaman yang sama seperti algoritma DL-AMR lanjutan lain, tetapi volum parameter hanya 1/12 daripada algoritma ini. Oleh itu, adalah berfaedah untuk menggunakan LDLSTM dalam sistem kekangan sumber.
Dong YI
PLA Strategic Support Force Information Engineering University
Di WU
PLA Strategic Support Force Information Engineering University
Tao HU
PLA Strategic Support Force Information Engineering University
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.
Salinan
Dong YI, Di WU, Tao HU, "A Lightweight Automatic Modulation Recognition Algorithm Based on Deep Learning" in IEICE TRANSACTIONS on Communications,
vol. E106-B, no. 4, pp. 367-373, April 2023, doi: 10.1587/transcom.2022EBP3087.
Abstract: Automatic modulation recognition (AMR) plays a critical role in modern communication systems. Owing to the recent advancements of deep learning (DL) techniques, the application of DL has been widely studied in AMR, and a large number of DL-AMR algorithms with high recognition rates have been developed. Most DL-AMR algorithm models have high recognition accuracy but have numerous parameters and are huge, complex models, which make them hard to deploy on resource-constrained platforms, such as satellite platforms. Some lightweight and low-complexity DL-AMR algorithm models also struggle to meet the accuracy requirements. Based on this, this paper proposes a lightweight and high-recognition-rate DL-AMR algorithm model called Lightweight Densely Connected Convolutional Network (DenseNet) Long Short-Term Memory network (LDLSTM). The model cascade of DenseNet and LSTM can achieve the same recognition accuracy as other advanced DL-AMR algorithms, but the parameter volume is only 1/12 that of these algorithms. Thus, it is advantageous to deploy LDLSTM in resource-constrained systems.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2022EBP3087/_p
Salinan
@ARTICLE{e106-b_4_367,
author={Dong YI, Di WU, Tao HU, },
journal={IEICE TRANSACTIONS on Communications},
title={A Lightweight Automatic Modulation Recognition Algorithm Based on Deep Learning},
year={2023},
volume={E106-B},
number={4},
pages={367-373},
abstract={Automatic modulation recognition (AMR) plays a critical role in modern communication systems. Owing to the recent advancements of deep learning (DL) techniques, the application of DL has been widely studied in AMR, and a large number of DL-AMR algorithms with high recognition rates have been developed. Most DL-AMR algorithm models have high recognition accuracy but have numerous parameters and are huge, complex models, which make them hard to deploy on resource-constrained platforms, such as satellite platforms. Some lightweight and low-complexity DL-AMR algorithm models also struggle to meet the accuracy requirements. Based on this, this paper proposes a lightweight and high-recognition-rate DL-AMR algorithm model called Lightweight Densely Connected Convolutional Network (DenseNet) Long Short-Term Memory network (LDLSTM). The model cascade of DenseNet and LSTM can achieve the same recognition accuracy as other advanced DL-AMR algorithms, but the parameter volume is only 1/12 that of these algorithms. Thus, it is advantageous to deploy LDLSTM in resource-constrained systems.},
keywords={},
doi={10.1587/transcom.2022EBP3087},
ISSN={1745-1345},
month={April},}
Salinan
TY - JOUR
TI - A Lightweight Automatic Modulation Recognition Algorithm Based on Deep Learning
T2 - IEICE TRANSACTIONS on Communications
SP - 367
EP - 373
AU - Dong YI
AU - Di WU
AU - Tao HU
PY - 2023
DO - 10.1587/transcom.2022EBP3087
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E106-B
IS - 4
JA - IEICE TRANSACTIONS on Communications
Y1 - April 2023
AB - Automatic modulation recognition (AMR) plays a critical role in modern communication systems. Owing to the recent advancements of deep learning (DL) techniques, the application of DL has been widely studied in AMR, and a large number of DL-AMR algorithms with high recognition rates have been developed. Most DL-AMR algorithm models have high recognition accuracy but have numerous parameters and are huge, complex models, which make them hard to deploy on resource-constrained platforms, such as satellite platforms. Some lightweight and low-complexity DL-AMR algorithm models also struggle to meet the accuracy requirements. Based on this, this paper proposes a lightweight and high-recognition-rate DL-AMR algorithm model called Lightweight Densely Connected Convolutional Network (DenseNet) Long Short-Term Memory network (LDLSTM). The model cascade of DenseNet and LSTM can achieve the same recognition accuracy as other advanced DL-AMR algorithms, but the parameter volume is only 1/12 that of these algorithms. Thus, it is advantageous to deploy LDLSTM in resource-constrained systems.
ER -