Joint Channel Equalization and Decoding with One Recurrent Neural Network

Yang Hu; Ling Zhao; Yue Hu

doi:10.1109/BMSB47279.2019.8971938

Joint Channel Equalization and Decoding with One Recurrent Neural Network

Yang Hu
, Ling Zhao
, Yue Hu

Beihang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

8 Scopus citations

Abstract

Channel equalization has been widely used to eliminate inter-symbol interference (ISI) and improve transmission performance in fading channel. In this paper, we propose a novel model of joint channel equalization and decoding based on recurrent neural network (RNN) in order to recover information messages interfered by channel distortion. By returning the output of decoder to the input of equalizer, an iterative equalizing and decoding process is achieved. Simulation over linear channel shows our method offers performance near that of maximum likelihood (ML) equalizer with knowledge of perfect channel state information (CSI). With less than 2/3 of the parameters, the proposed model has more than 0.5 dB gain over the CNN + NND-Joint model (and three other models) over nonlinear channel.

Original language	English
Title of host publication	2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019
Publisher	IEEE Computer Society
ISBN (Electronic)	9781728121505
DOIs	https://doi.org/10.1109/BMSB47279.2019.8971938
State	Published - Jun 2019
Event	2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019 - Jeju, Korea, Republic of Duration: 5 Jun 2019 → 7 Jun 2019

Publication series

Name	IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB
Volume	2019-June
ISSN (Print)	2155-5044
ISSN (Electronic)	2155-5052

Conference

Conference	2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019
Country/Territory	Korea, Republic of
City	Jeju
Period	5/06/19 → 7/06/19

Keywords

channel coding
channel estimation and equalization
recurrent neural network (RNN)

Access to Document

10.1109/BMSB47279.2019.8971938

Cite this

Hu, Y., Zhao, L., & Hu, Y. (2019). Joint Channel Equalization and Decoding with One Recurrent Neural Network. In 2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019 Article 8971938 (IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB; Vol. 2019-June). IEEE Computer Society. https://doi.org/10.1109/BMSB47279.2019.8971938

@inproceedings{97ff7935bd7e4f639276f0a4f1ec58a9,

title = "Joint Channel Equalization and Decoding with One Recurrent Neural Network",

abstract = "Channel equalization has been widely used to eliminate inter-symbol interference (ISI) and improve transmission performance in fading channel. In this paper, we propose a novel model of joint channel equalization and decoding based on recurrent neural network (RNN) in order to recover information messages interfered by channel distortion. By returning the output of decoder to the input of equalizer, an iterative equalizing and decoding process is achieved. Simulation over linear channel shows our method offers performance near that of maximum likelihood (ML) equalizer with knowledge of perfect channel state information (CSI). With less than 2/3 of the parameters, the proposed model has more than 0.5 dB gain over the CNN + NND-Joint model (and three other models) over nonlinear channel.",

keywords = "channel coding, channel estimation and equalization, recurrent neural network (RNN)",

author = "Yang Hu and Ling Zhao and Yue Hu",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019 ; Conference date: 05-06-2019 Through 07-06-2019",

year = "2019",

month = jun,

doi = "10.1109/BMSB47279.2019.8971938",

language = "英语",

series = "IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB",

publisher = "IEEE Computer Society",

booktitle = "2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019",

address = "美国",

}

Hu, Y , Zhao, L & Hu, Y 2019, Joint Channel Equalization and Decoding with One Recurrent Neural Network. in 2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019., 8971938, IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB, vol. 2019-June, IEEE Computer Society, 2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019, Jeju, Korea, Republic of, 5/06/19. https://doi.org/10.1109/BMSB47279.2019.8971938

Joint Channel Equalization and Decoding with One Recurrent Neural Network. / Hu, Yang ; Zhao, Ling; Hu, Yue.
2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019. IEEE Computer Society, 2019. 8971938 (IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB; Vol. 2019-June).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Joint Channel Equalization and Decoding with One Recurrent Neural Network

AU - Hu, Yang

AU - Zhao, Ling

AU - Hu, Yue

PY - 2019/6

Y1 - 2019/6

N2 - Channel equalization has been widely used to eliminate inter-symbol interference (ISI) and improve transmission performance in fading channel. In this paper, we propose a novel model of joint channel equalization and decoding based on recurrent neural network (RNN) in order to recover information messages interfered by channel distortion. By returning the output of decoder to the input of equalizer, an iterative equalizing and decoding process is achieved. Simulation over linear channel shows our method offers performance near that of maximum likelihood (ML) equalizer with knowledge of perfect channel state information (CSI). With less than 2/3 of the parameters, the proposed model has more than 0.5 dB gain over the CNN + NND-Joint model (and three other models) over nonlinear channel.

AB - Channel equalization has been widely used to eliminate inter-symbol interference (ISI) and improve transmission performance in fading channel. In this paper, we propose a novel model of joint channel equalization and decoding based on recurrent neural network (RNN) in order to recover information messages interfered by channel distortion. By returning the output of decoder to the input of equalizer, an iterative equalizing and decoding process is achieved. Simulation over linear channel shows our method offers performance near that of maximum likelihood (ML) equalizer with knowledge of perfect channel state information (CSI). With less than 2/3 of the parameters, the proposed model has more than 0.5 dB gain over the CNN + NND-Joint model (and three other models) over nonlinear channel.

KW - channel coding

KW - channel estimation and equalization

KW - recurrent neural network (RNN)

UR - https://www.scopus.com/pages/publications/85079358002

U2 - 10.1109/BMSB47279.2019.8971938

DO - 10.1109/BMSB47279.2019.8971938

M3 - 会议稿件

AN - SCOPUS:85079358002

T3 - IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB

BT - 2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019

PB - IEEE Computer Society

T2 - 2019 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB 2019

Y2 - 5 June 2019 through 7 June 2019

ER -

Joint Channel Equalization and Decoding with One Recurrent Neural Network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this