Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission

The performance of optical mode division multiplexing (MDM) is affected by intersymbol interference (ISI) from nonlinear channel impairments arising from higherorder mode coupling and modal dispersion in multimode fiber. However, the existing MDM equalization algorithms can only mitigate the linear...

Full description

Saved in:

Bibliographic Details
Main Author:	Noori, Awab
Format:	Thesis
Language:	eng
Published:	2017
Subjects:	QA76.76 Fuzzy System. T Technology (General)
Online Access:	https://etd.uum.edu.my/10271/1/s817063_01.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-uum-etd.10271
record_format	uketd_dc
spelling	my-uum-etd.102712023-02-01T00:06:00Z Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission 2017 Noori, Awab Sop Chit, Suwannit Chareen Amphawan, Angela College of Arts and Sciences (CAS) College of Arts and Sciences (CAS) QA76.76 Fuzzy System. T Technology (General) The performance of optical mode division multiplexing (MDM) is affected by intersymbol interference (ISI) from nonlinear channel impairments arising from higherorder mode coupling and modal dispersion in multimode fiber. However, the existing MDM equalization algorithms can only mitigate the linear distortion, but they cannot address nonlinear distortion in the signal accurately. Therefore, there is a need to explore how ISI can be mitigated to recover the transmitted signal. This research aims to control the broadening of the MDM signal and minimize the undesirable distortion among channels in MMF by signal reshaping at the receiver. A dynamic evolving neural fuzzy inference system (DENFIS) equalization scheme has been used to achieve this objective. This research was conducted through a few steps commencing with modelling the MDM system in Optsim and collecting the data. Then, the signal reshaping parameters were determined. After that, DENFIS equalization, least mean square (LMS) and recursive least squares (RLS) equalizations were implemented and evaluated. Results illustrated that nonlinear DENFIS equalization scheme can improve MDM signal at a higher accuracy than previous linear equalization schemes. DENFIS equalization demonstrates better signal reshaping accuracy with an average root mean square error (RMSE) of 0.0338 and outperformed linear LMS and RLS equalization schemes with high average RMSE values of 0.101 and 0.1914 respectively. The reduced RMSE implies that DENFIS equalization scheme mitigates ISI more effectively in a nonlinear channel. This effect can hasten data transmission rates in MDM. Moreover, the successful offline implementation of DENFIS equalization in MDM encourages future online implementation of DENFIS equalization in embedded optical systems. 2017 Thesis https://etd.uum.edu.my/10271/ https://etd.uum.edu.my/10271/1/s817063_01.pdf text eng public other masters Universiti Utara Malaysia
institution	Universiti Utara Malaysia
collection	UUM ETD
language	eng
advisor	Sop Chit, Suwannit Chareen Amphawan, Angela
topic	QA76.76 Fuzzy System. T Technology (General)
spellingShingle	QA76.76 Fuzzy System. T Technology (General) Noori, Awab Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission
description	The performance of optical mode division multiplexing (MDM) is affected by intersymbol interference (ISI) from nonlinear channel impairments arising from higherorder mode coupling and modal dispersion in multimode fiber. However, the existing MDM equalization algorithms can only mitigate the linear distortion, but they cannot address nonlinear distortion in the signal accurately. Therefore, there is a need to explore how ISI can be mitigated to recover the transmitted signal. This research aims to control the broadening of the MDM signal and minimize the undesirable distortion among channels in MMF by signal reshaping at the receiver. A dynamic evolving neural fuzzy inference system (DENFIS) equalization scheme has been used to achieve this objective. This research was conducted through a few steps commencing with modelling the MDM system in Optsim and collecting the data. Then, the signal reshaping parameters were determined. After that, DENFIS equalization, least mean square (LMS) and recursive least squares (RLS) equalizations were implemented and evaluated. Results illustrated that nonlinear DENFIS equalization scheme can improve MDM signal at a higher accuracy than previous linear equalization schemes. DENFIS equalization demonstrates better signal reshaping accuracy with an average root mean square error (RMSE) of 0.0338 and outperformed linear LMS and RLS equalization schemes with high average RMSE values of 0.101 and 0.1914 respectively. The reduced RMSE implies that DENFIS equalization scheme mitigates ISI more effectively in a nonlinear channel. This effect can hasten data transmission rates in MDM. Moreover, the successful offline implementation of DENFIS equalization in MDM encourages future online implementation of DENFIS equalization in embedded optical systems.
format	Thesis
qualification_name	other
qualification_level	Master's degree
author	Noori, Awab
author_facet	Noori, Awab
author_sort	Noori, Awab
title	Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission
title_short	Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission
title_full	Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission
title_fullStr	Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission
title_full_unstemmed	Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission
title_sort	dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission
granting_institution	Universiti Utara Malaysia
granting_department	College of Arts and Sciences (CAS)
publishDate	2017
url	https://etd.uum.edu.my/10271/1/s817063_01.pdf
_version_	1776103779523887104

Dynamic evolving neural fuzzy inference system equalization scheme in mode division multiplexing for optical fiber transmission

Similar Items