Joint Statistics and MLSD in Filtered Incoherent High-Speed Fiber-Optic Communications

Author

Alic, Nikola ; Karlsson, Magnus ; Sköld, M. ; Milenkovic, Olgica ; Andrekson, P.A. ; Radic, Stojan

Author_Institution

Dept. of Electr. & Comput. Eng., Univ. of California San Diego, La Jolla, CA, USA

Volume

28

Issue

10

fYear

2010

fDate

5/15/2010 12:00:00 AM

Firstpage

1564

Lastpage

1572

Abstract

In this paper, signal statistics and their utilization for detection in narrowly filtered equalized high-speed fiber-optic communications are investigated experimentally. Tradeoffs between log-likelihood metric applications and oversampling are covered in detail. It is, for the first time, demonstrated that performance loss in bandwidth-limited systems can be nearly fully recovered (to within 0.5 dB) by taking advantage of bandwidth-limitation-induced noise correlations and oversampling.

Keywords

error statistics; filtering theory; maximum likelihood sequence estimation; optical fibre communication; MLSD; Viterbi algorithm; bandwidth-limitation-induced noise correlations; bandwidth-limited systems; bit-error ratio minimization; colored noise; filtered incoherent high-speed fiber-optic communications; joint statistics; log-likelihood metric; maximum-likelihood sequence detection; signal statistics; Colored noise; Viterbi algorithm; fiber-optic communications; maximum-likelihood sequence detection;

fLanguage

English

Journal_Title

Lightwave Technology, Journal of

Publisher

ieee

ISSN

0733-8724

Type

jour

DOI

10.1109/JLT.2010.2043054

Filename

5418986