City University of Hong Kong

CityU Institutional Repository >
3_CityU Electronic Theses and Dissertations >
ETD - Dept. of Computer Science  >
CS - Doctor of Philosophy  >

Please use this identifier to cite or link to this item:

Title: Efficient bandwidth utilization in optical WDM networks
Other Titles: Guang bo fen duo lu fu yong wang luo zhong guan yu ti gao dai kuan li yong lü wen ti de yan jiu
Authors: Li, Yan ( 李燕)
Department: Department of Computer Science
Degree: Doctor of Philosophy
Issue Date: 2011
Publisher: City University of Hong Kong
Subjects: Wavelength division multiplexing.
Broadband communication systems.
Optical communications.
Notes: CityU Call Number: TK5103.592.W38 L54 2011
viii, 105 leaves : ill. 30 cm.
Thesis (Ph.D.)--City University of Hong Kong, 2011.
Includes bibliographical references (leaves 96-105)
Type: thesis
Abstract: Optical wavelength division multiplexing (WDM) technology has been considered as the basis of the next generation network (NGN) which consists of the core network and the access network. WDM significantly increases capacity of optical fibers by exploiting multiple wavelength channels on each fiber. In the past decades, great efforts have been made on evaluating the bandwidth utilization in WDM networks. However, in WDM core networks, the mismatch between the high bandwidth per wavelength can provide and relatively low bandwidth requirement of a single connection limits the satisfaction of ever-changing service requirements by using commodity optical devices. The similar problem also exists in WDM access networks. To achieve a perfect balance between provision of enormous access bandwidth and cost-effectiveness of infrastructure deployment, integrated Fiber-Wireless (FiWi) access networks which employ optical WDM access network as back end and wireless access network as front end are gaining rapid popularity as a promising candidate for future access networks. In FiWi access networks, the low throughput in wireless front end due to interference limits the utilization of the huge bandwidth provided by the optical WDM access networks. In this dissertation, we study how to fully utilize the huge bandwidth which is introduced by WDM technology in both WDM core networks and WDM access networks, such that desirable network performance can be achieved. The main research issues and contributions are summarized as follows: Firstly, we study multicast routing in light-trail WDM core networks. Recently, light-trail is becoming an appealing architecture for WDM core networks which have been considered as promising candidate of the NGN. Light-trail can inherently support multicast given its bus nature. In our work, we study how to use minimum number of light-trails to formulate multicast tree for supporting the given multicast session, such that optical bandwidth can be efficiently utilized. The problem for general light-trail WDM networks is proved to be NP-hard. Two auxiliary graphs will be proposed to transform the problem into minimum Steiner tree problem that many effective algorithms can be applied. We then show that the same problem in light-trail WDM ring networks can be solved in polynomial time. The simulations show the effectiveness of our work. Secondly, we propose an adaptive intentional connection rerouting scheme for traffic grooming in WDM core networks. In WDM core networks, bandwidth utilization is limited by the mismatch between the high bandwidth that one wavelength can provide and the relatively low bandwidth request of a single connection. Traffic grooming which grooms multiple connections onto one wavelength is a desirable technique to improve resource utilization. However, with unpredictable traffic demand, the newly arrived request may not be accommodated using remaining resources. Rerouting which reroutes existing connections such that the newly arrived request can be accommodated is an effective approach to further improve network throughput. In this work, on the basis of biological attractor selection approach, we study adaptive intentional connection rerouting which intentionally triggers rerouting algorithm according to network status, with the aim to reserve maximum resources for future connections. Then, the network blocking probability can be reduced within an appropriate number of rerouting by efficiently using the huge optical bandwidth. The proposed rerouting scheme can also improve system robustness given its adaptability. The simulations demonstrate the effectiveness of our proposed adaptive rerouting approach. Finally, we address how to efficiently utilize the high bandwidth provided by optical WDM access network which is employed as back end of FiWi access networks. FiWi access networks provide a powerful platform to improve the throughput of peer-to-peer communication by enabling traffic to be sent from the source wireless client to an ingress optical network unit (ONU), then to the egress ONU close to the destination wireless client, and finally delivered to the destination wireless client. With the support of direct inter-ONU communication in WDM back end, the huge bandwidth provided by WDM access networks can be efficiently used to improve the throughput of peer-to-peer communication in FiWi access networks. In this study, we propose a novel WDM passive optical network (PON) architecture supporting direct inter-ONU communication, a wavelength assignment algorithm in the proposed WDM PON, a corresponding decentralized dynamic bandwidth allocation (DBA) protocol for inter-ONU communication and an algorithm to dynamically select egress ONU. The complexity of the proposed architecture is analyzed and compared with other alternatives, and the efficiency of the proposed system is validated by the simulations. Key words: WDM Network Core Access Bandwidth Utilization
Online Catalog Link:
Appears in Collections:CS - Doctor of Philosophy

Files in This Item:

File Description SizeFormat
abstract.html132 BHTMLView/Open
fulltext.html132 BHTMLView/Open

Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0!
DSpace Software © 2013 CityU Library - Send feedback to Library Systems
Privacy Policy · Copyright · Disclaimer