Probabilistically Shaped 8-PAM Suitable for Data Centers Communication

Research output: Chapter in Book/Report/Conference proceedingConference contribution

8 Scopus citations

Abstract

We study non-uniform 8-PAM signaling schemes, based on Maxwell-Boltzmann and exponential source distributions, as candidates suitable for data centers' communications. We evaluate mutual information of these schemes against corresponding uniform 8-PAM scheme, for different parameters of source distributions. The properly chosen variable parameters for different SNR-regions allows us to closely approach Shannon limit, and to significantly outperform the corresponding uniform counterpart. We demonstrate that LPDC-coded 8-PAM with exponential distribution outperforms the LDPC-coded 8-PAM with uniform distribution of the same achievable information rate by 1.8 dB, and this improvement is comparable to that obtained from mutual information calculations.

Original languageEnglish (US)
Title of host publication2018 20th International Conference on Transparent Optical Networks, ICTON 2018
PublisherIEEE Computer Society
Volume2018-July
ISBN (Electronic)9781538666043
DOIs
StatePublished - Sep 26 2018
Event20th International Conference on Transparent Optical Networks, ICTON 2018 - Bucharest, Romania
Duration: Jul 1 2018Jul 5 2018

Other

Other20th International Conference on Transparent Optical Networks, ICTON 2018
CountryRomania
CityBucharest
Period7/1/187/5/18

Keywords

  • 8-PAM
  • distribution matching
  • exponential distribution
  • LDPC coding.
  • probabilistic shaping

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials

Fingerprint Dive into the research topics of 'Probabilistically Shaped 8-PAM Suitable for Data Centers Communication'. Together they form a unique fingerprint.

  • Cite this

    Han, X., & Djordjevic, I. B. (2018). Probabilistically Shaped 8-PAM Suitable for Data Centers Communication. In 2018 20th International Conference on Transparent Optical Networks, ICTON 2018 (Vol. 2018-July). [8473738] IEEE Computer Society. https://doi.org/10.1109/ICTON.2018.8473738