Enhanced Multi-Level Signal Recovery in Mobile Fronthaul Network Using DNN Decoder

Qi Zhou; Feng Lu; Mu Xu; Peng Chun Peng; Siming Liu; Shuyi Shen; Rui Zhang; Shuang Yao; Jeffrey Finkelstein; Gee Kung Chang

doi:10.1109/LPT.2018.2852601

Enhanced Multi-Level Signal Recovery in Mobile Fronthaul Network Using DNN Decoder

Qi Zhou
, Feng Lu
, Mu Xu
, Peng Chun Peng
, Siming Liu
, Shuyi Shen
, Rui Zhang
, Shuang Yao
, Jeffrey Finkelstein
, Gee Kung Chang

Department of Electrical Engineering

Georgia Institute of Technology
National Taipei University of Technology
Cox Communications

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

26 Scopus citations

Abstract

A deep neural network (DNN) decoder which combines the function of both equalization and decoding is proposed and experimentally demonstrated for mobile fronthaul (MFH) transmission. This DNN consists of one input layer, one output layer, and several hidden layers. Adamax algorithm is implemented for finding the global minima while dropout mechanism and early stopping are utilized to avoid overfitting. The DNN accepts the received samples as the input and output the decoded samples directly to recover the transmitted samples. Using this DNN decoder, a record-high data-rate transmission-distance product at 1800-Gb/s $\cdot $ km based on the directly modulated laser (DML) with intensity-modulation direct-detection is obtained. Besides, the PAM8 modulation with 3-bps/Hz spectral efficiency is implemented. Due to the smaller source spectrum bandwidth compared with traditionally widely used PAM4 and OOK at a certain data rate, the power fading limited transmission distance is extended by 1.5 and three times, respectively. The DML is a low-cost device and this simple transmission system eliminates the need of single-side-band modulation or dispersion compensation, which makes this system an ideal candidate for a low-cost enhanced MFH network.

Original language	English
Article number	8401905
Pages (from-to)	1511-1514
Number of pages	4
Journal	IEEE Photonics Technology Letters
Volume	30
Issue number	17
DOIs	https://doi.org/10.1109/LPT.2018.2852601
State	Published - Sep 1 2018

Keywords

deep neural network
digital signal processing
Mobile fronthaul
optical fiber communication
optical modulation

Access to Document

10.1109/LPT.2018.2852601

Cite this

@article{7308da911e3b457e9aa196f3a3b2d856,

title = "Enhanced Multi-Level Signal Recovery in Mobile Fronthaul Network Using DNN Decoder",

abstract = "A deep neural network (DNN) decoder which combines the function of both equalization and decoding is proposed and experimentally demonstrated for mobile fronthaul (MFH) transmission. This DNN consists of one input layer, one output layer, and several hidden layers. Adamax algorithm is implemented for finding the global minima while dropout mechanism and early stopping are utilized to avoid overfitting. The DNN accepts the received samples as the input and output the decoded samples directly to recover the transmitted samples. Using this DNN decoder, a record-high data-rate transmission-distance product at 1800-Gb/s \$\textbackslash{}cdot \$ km based on the directly modulated laser (DML) with intensity-modulation direct-detection is obtained. Besides, the PAM8 modulation with 3-bps/Hz spectral efficiency is implemented. Due to the smaller source spectrum bandwidth compared with traditionally widely used PAM4 and OOK at a certain data rate, the power fading limited transmission distance is extended by 1.5 and three times, respectively. The DML is a low-cost device and this simple transmission system eliminates the need of single-side-band modulation or dispersion compensation, which makes this system an ideal candidate for a low-cost enhanced MFH network.",

keywords = "deep neural network, digital signal processing, Mobile fronthaul, optical fiber communication, optical modulation",

author = "Qi Zhou and Feng Lu and Mu Xu and Peng, \{Peng Chun\} and Siming Liu and Shuyi Shen and Rui Zhang and Shuang Yao and Jeffrey Finkelstein and Chang, \{Gee Kung\}",

note = "Publisher Copyright: {\textcopyright} 1989-2012 IEEE.",

year = "2018",

month = sep,

day = "1",

doi = "10.1109/LPT.2018.2852601",

language = "English",

volume = "30",

pages = "1511--1514",

journal = "IEEE Photonics Technology Letters",

issn = "1041-1135",

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

number = "17",

}

TY - JOUR

T1 - Enhanced Multi-Level Signal Recovery in Mobile Fronthaul Network Using DNN Decoder

AU - Zhou, Qi

AU - Lu, Feng

AU - Xu, Mu

AU - Peng, Peng Chun

AU - Liu, Siming

AU - Shen, Shuyi

AU - Zhang, Rui

AU - Yao, Shuang

AU - Finkelstein, Jeffrey

AU - Chang, Gee Kung

PY - 2018/9/1

Y1 - 2018/9/1

N2 - A deep neural network (DNN) decoder which combines the function of both equalization and decoding is proposed and experimentally demonstrated for mobile fronthaul (MFH) transmission. This DNN consists of one input layer, one output layer, and several hidden layers. Adamax algorithm is implemented for finding the global minima while dropout mechanism and early stopping are utilized to avoid overfitting. The DNN accepts the received samples as the input and output the decoded samples directly to recover the transmitted samples. Using this DNN decoder, a record-high data-rate transmission-distance product at 1800-Gb/s $\cdot $ km based on the directly modulated laser (DML) with intensity-modulation direct-detection is obtained. Besides, the PAM8 modulation with 3-bps/Hz spectral efficiency is implemented. Due to the smaller source spectrum bandwidth compared with traditionally widely used PAM4 and OOK at a certain data rate, the power fading limited transmission distance is extended by 1.5 and three times, respectively. The DML is a low-cost device and this simple transmission system eliminates the need of single-side-band modulation or dispersion compensation, which makes this system an ideal candidate for a low-cost enhanced MFH network.

AB - A deep neural network (DNN) decoder which combines the function of both equalization and decoding is proposed and experimentally demonstrated for mobile fronthaul (MFH) transmission. This DNN consists of one input layer, one output layer, and several hidden layers. Adamax algorithm is implemented for finding the global minima while dropout mechanism and early stopping are utilized to avoid overfitting. The DNN accepts the received samples as the input and output the decoded samples directly to recover the transmitted samples. Using this DNN decoder, a record-high data-rate transmission-distance product at 1800-Gb/s $\cdot $ km based on the directly modulated laser (DML) with intensity-modulation direct-detection is obtained. Besides, the PAM8 modulation with 3-bps/Hz spectral efficiency is implemented. Due to the smaller source spectrum bandwidth compared with traditionally widely used PAM4 and OOK at a certain data rate, the power fading limited transmission distance is extended by 1.5 and three times, respectively. The DML is a low-cost device and this simple transmission system eliminates the need of single-side-band modulation or dispersion compensation, which makes this system an ideal candidate for a low-cost enhanced MFH network.

KW - deep neural network

KW - digital signal processing

KW - Mobile fronthaul

KW - optical fiber communication

KW - optical modulation

UR - https://www.scopus.com/pages/publications/85049327024

U2 - 10.1109/LPT.2018.2852601

DO - 10.1109/LPT.2018.2852601

M3 - Article

AN - SCOPUS:85049327024

SN - 1041-1135

VL - 30

SP - 1511

EP - 1514

JO - IEEE Photonics Technology Letters

JF - IEEE Photonics Technology Letters

IS - 17

M1 - 8401905

ER -

Enhanced Multi-Level Signal Recovery in Mobile Fronthaul Network Using DNN Decoder

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this