A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems

A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems

Deep learning has been applied in physical-layer communications systems in recent years and has demonstrated fascinating results that were comparable or even better than human expert systems. In this paper, a novel convolutional neural networks (CNNs)-based autoencoder communication system is propos...

Full description

Bibliographic Details
Main Authors: Nan Wu, Xudong Wang, Bin Lin, Kaiyao Zhang
Format: Article
Language:English
Published: IEEE 2019-01-01
Series:IEEE Access
Subjects:
Convolutional neural network
end-to-end learning
autoencoder
communication systems
Online Access:https://ieeexplore.ieee.org/document/8755977/
id doaj-9de77395be5d444a83f11ac514306687
record_format Article
spelling doaj-9de77395be5d444a83f11ac5143066872021-04-05T17:21:46ZengIEEEIEEE Access2169-35362019-01-01711019711020410.1109/ACCESS.2019.29268438755977A CNN-Based End-to-End Learning Framework Toward Intelligent Communication SystemsNan Wu0Xudong Wang1Bin Lin2Kaiyao Zhang3https://orcid.org/0000-0003-4396-673XSchool of Information Science and Technology, Dalian Maritime University, Dalian, ChinaSchool of Information Science and Technology, Dalian Maritime University, Dalian, ChinaSchool of Information Science and Technology, Dalian Maritime University, Dalian, ChinaSchool of Information Science and Technology, Dalian Maritime University, Dalian, ChinaDeep learning has been applied in physical-layer communications systems in recent years and has demonstrated fascinating results that were comparable or even better than human expert systems. In this paper, a novel convolutional neural networks (CNNs)-based autoencoder communication system is proposed, which can work intelligently with arbitrary block length, can support different throughput and can operate under AWGN and Rayleigh fading channels as well as deviations from AWGN environments. The proposed generalized communication system is comprised of carefully designed convolutional neural layers and, hence, inherits CNN's breakthrough characteristics, such as generalization, feature learning, classification, and fast training convergence. On the other hand, the end-to-end architecture jointly performs the tasks of encoding/decoding and modulation/demodulation. Finally, we provide the numerous simulation results of the learned system in order to illustrate its generalization capability under various system conditions.https://ieeexplore.ieee.org/document/8755977/Convolutional neural networkend-to-end learningautoencodercommunication systems
collection DOAJ
language English
format Article
sources DOAJ
author Nan Wu
Xudong Wang
Bin Lin
Kaiyao Zhang
spellingShingle Nan Wu
Xudong Wang
Bin Lin
Kaiyao Zhang
A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems
IEEE Access
Convolutional neural network
end-to-end learning
autoencoder
communication systems
author_facet Nan Wu
Xudong Wang
Bin Lin
Kaiyao Zhang
author_sort Nan Wu
title A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems
title_short A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems
title_full A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems
title_fullStr A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems
title_full_unstemmed A CNN-Based End-to-End Learning Framework Toward Intelligent Communication Systems
title_sort cnn-based end-to-end learning framework toward intelligent communication systems
publisher IEEE
series IEEE Access
issn 2169-3536
publishDate 2019-01-01
description Deep learning has been applied in physical-layer communications systems in recent years and has demonstrated fascinating results that were comparable or even better than human expert systems. In this paper, a novel convolutional neural networks (CNNs)-based autoencoder communication system is proposed, which can work intelligently with arbitrary block length, can support different throughput and can operate under AWGN and Rayleigh fading channels as well as deviations from AWGN environments. The proposed generalized communication system is comprised of carefully designed convolutional neural layers and, hence, inherits CNN's breakthrough characteristics, such as generalization, feature learning, classification, and fast training convergence. On the other hand, the end-to-end architecture jointly performs the tasks of encoding/decoding and modulation/demodulation. Finally, we provide the numerous simulation results of the learned system in order to illustrate its generalization capability under various system conditions.
topic Convolutional neural network
end-to-end learning
autoencoder
communication systems
url https://ieeexplore.ieee.org/document/8755977/
work_keys_str_mv AT nanwu acnnbasedendtoendlearningframeworktowardintelligentcommunicationsystems
AT xudongwang acnnbasedendtoendlearningframeworktowardintelligentcommunicationsystems
AT binlin acnnbasedendtoendlearningframeworktowardintelligentcommunicationsystems
AT kaiyaozhang acnnbasedendtoendlearningframeworktowardintelligentcommunicationsystems
AT nanwu cnnbasedendtoendlearningframeworktowardintelligentcommunicationsystems
AT xudongwang cnnbasedendtoendlearningframeworktowardintelligentcommunicationsystems
AT binlin cnnbasedendtoendlearningframeworktowardintelligentcommunicationsystems
AT kaiyaozhang cnnbasedendtoendlearningframeworktowardintelligentcommunicationsystems
_version_ 1721539822010499072

Similar Items