Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems

Chun Yen Chuang; Li Chun Liu; Chia Chien Wei; Jun Jie Liu; Lindor Henrickson; Chih Lin Wang; Young Kai Chen; Jye-Hong Chen

doi:10.1109/ECOC.2018.8535400

Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems

Chun Yen Chuang, Li Chun Liu, Chia Chien Wei^*, Jun Jie Liu, Lindor Henrickson, Chih Lin Wang, Young Kai Chen, Jye-Hong Chen

^*此作品的通信作者

光電工程學系

研究成果: Conference contribution › 同行評審

16 引文斯高帕斯（Scopus）

摘要

Our investigation demonstrates that training patterns generated by Mersenne Twister pseudo-random number generator have potential to overcome the pattern effect. The resulting trained networks are capable of compensating the actual optical communication system response rather than merely predicting training signals.

原文	English
主出版物標題	2018 European Conference on Optical Communication, ECOC 2018
發行者	Institute of Electrical and Electronics Engineers Inc.
ISBN（電子）	9781538648629
DOIs	https://doi.org/10.1109/ECOC.2018.8535400
出版狀態	Published - 14 11月 2018
事件	2018 European Conference on Optical Communication, ECOC 2018 - Rome, Italy 持續時間: 23 9月 2018 → 27 9月 2018

出版系列

名字	European Conference on Optical Communication, ECOC
卷	2018-September

Conference

Conference	2018 European Conference on Optical Communication, ECOC 2018
國家/地區	Italy
城市	Rome
期間	23/09/18 → 27/09/18

存取文件

10.1109/ECOC.2018.8535400

其它文件與鏈接

Link to publication in Scopus

引用此

Chuang, C. Y., Liu, L. C., Wei, C. C., Liu, J. J., Henrickson, L., Wang, C. L., Chen, Y. K., & Chen, J-H. (2018). Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems. 於 2018 European Conference on Optical Communication, ECOC 2018 Article 8535400 (European Conference on Optical Communication, ECOC; 卷 2018-September). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ECOC.2018.8535400

@inproceedings{db64bee9288e4411bf8c2b54da187726,

title = "Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems",

abstract = "Our investigation demonstrates that training patterns generated by Mersenne Twister pseudo-random number generator have potential to overcome the pattern effect. The resulting trained networks are capable of compensating the actual optical communication system response rather than merely predicting training signals.",

author = "Chuang, {Chun Yen} and Liu, {Li Chun} and Wei, {Chia Chien} and Liu, {Jun Jie} and Lindor Henrickson and Wang, {Chih Lin} and Chen, {Young Kai} and Jye-Hong Chen",

year = "2018",

month = nov,

day = "14",

doi = "10.1109/ECOC.2018.8535400",

language = "English",

series = "European Conference on Optical Communication, ECOC",

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

booktitle = "2018 European Conference on Optical Communication, ECOC 2018",

address = "United States",

note = "2018 European Conference on Optical Communication, ECOC 2018 ; Conference date: 23-09-2018 Through 27-09-2018",

}

Chuang, CY, Liu, LC, Wei, CC, Liu, JJ, Henrickson, L, Wang, CL, Chen, YK & Chen, J-H 2018, Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems. 於 2018 European Conference on Optical Communication, ECOC 2018., 8535400, European Conference on Optical Communication, ECOC, 卷 2018-September, Institute of Electrical and Electronics Engineers Inc., 2018 European Conference on Optical Communication, ECOC 2018, Rome, Italy, 23/09/18. https://doi.org/10.1109/ECOC.2018.8535400

Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems. / Chuang, Chun Yen; Liu, Li Chun; Wei, Chia Chien 等.
2018 European Conference on Optical Communication, ECOC 2018. Institute of Electrical and Electronics Engineers Inc., 2018. 8535400 (European Conference on Optical Communication, ECOC; 卷 2018-September).

研究成果: Conference contribution › 同行評審

TY - GEN

T1 - Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems

AU - Chuang, Chun Yen

AU - Liu, Li Chun

AU - Wei, Chia Chien

AU - Liu, Jun Jie

AU - Henrickson, Lindor

AU - Wang, Chih Lin

AU - Chen, Young Kai

AU - Chen, Jye-Hong

PY - 2018/11/14

Y1 - 2018/11/14

N2 - Our investigation demonstrates that training patterns generated by Mersenne Twister pseudo-random number generator have potential to overcome the pattern effect. The resulting trained networks are capable of compensating the actual optical communication system response rather than merely predicting training signals.

AB - Our investigation demonstrates that training patterns generated by Mersenne Twister pseudo-random number generator have potential to overcome the pattern effect. The resulting trained networks are capable of compensating the actual optical communication system response rather than merely predicting training signals.

UR - http://www.scopus.com/inward/record.url?scp=85063191413&partnerID=8YFLogxK

U2 - 10.1109/ECOC.2018.8535400

DO - 10.1109/ECOC.2018.8535400

M3 - Conference contribution

AN - SCOPUS:85063191413

T3 - European Conference on Optical Communication, ECOC

BT - 2018 European Conference on Optical Communication, ECOC 2018

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2018 European Conference on Optical Communication, ECOC 2018

Y2 - 23 September 2018 through 27 September 2018

ER -

Chuang CY, Liu LC, Wei CC, Liu JJ, Henrickson L, Wang CL 等. Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems. 於 2018 European Conference on Optical Communication, ECOC 2018. Institute of Electrical and Electronics Engineers Inc. 2018. 8535400. (European Conference on Optical Communication, ECOC). doi: 10.1109/ECOC.2018.8535400

Study of Training Patterns for Employing Deep Neural Networks in Optical Communication Systems

摘要

出版系列

Conference

存取文件

其它文件與鏈接

指紋

引用此