ASIC Implementation of Time-Domain Digital Backpropagation with Deep-Learned Chromatic Dispersion Filters
Paper in proceeding, 2018
We consider time-domain digital backpropagation with chromatic dispersion filters jointly optimized and quantized using machine-learning techniques. Compared to the baseline implementations, we show improved BER performance and >40% power dissipation reductions in 28-nm CMOS.
Author
Christoffer Fougstedt
Chalmers, Computer Science and Engineering (Chalmers), Computer Engineering (Chalmers)
Christian Häger
Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks
Duke University
Lars Svensson
Chalmers, Computer Science and Engineering (Chalmers), Computer Engineering (Chalmers)
Henry D. Pfister
Duke University
Per Larsson-Edefors
Chalmers, Computer Science and Engineering (Chalmers), Computer Engineering (Chalmers)
Published in
European Conference on Optical Communication, ECOC
Vol. 2018-September art. no 8535430
978-153864862-9 (ISBN)
Conference
2018 European Conference on Optical Communication, ECOC 2018
Rome, Italy, 2018-09-22 - 2018-09-26
Rome, Italy, 2018-09-22 - 2018-09-26
Research Project(s)
Coding for terabit-per-second fiber-optical communications (TERA)
European Commission (EC) (EC/H2020/749798), 2017-01-01 -- 2019-12-31.
Energy-efficient optical fibre communication
Knut and Alice Wallenberg Foundation (KAW2013.0021), 2014-07-01 -- 2019-06-30.
Categorizing
Areas of Advance
Information and Communication Technology
Driving Forces
Sustainable development
Subject Categories (SSIF 2011)
Signal Processing
Computer Systems
Other Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
DOI
10.1109/ECOC.2018.8535430