Optical WDM Networks E-Book

Devi Chadha

• Herausgeber: John Wiley & Sons
• Kategorie: Wissenschaft und neue Technologien
• Serie: Wiley - IEEE
• Sprache: Englisch

Seitenzahl: 682

Veröffentlichungsjahr: 2019

Beschreibung

Provides a comprehensive and updated account of WDM optical network systems Optical networking has advanced considerably since 2010. A host of new technologies and applications has brought a significant change in optical networks, migrating it towards an all-optical network. This book places great emphasis on the network concepts, technology, and methodologies that will stand the test of time and also help in understanding and developing advanced optical network systems. The first part of Optical WDM Networks: From Static to Elastic Networks provides a qualitative foundation for what follows--presenting an overview of optical networking, the different network architectures, basic concepts, and a high-level view of the different network structures considered in subsequent chapters. It offers a survey of enabling technologies and the hardware devices in the physical layer, followed by a more detailed picture of the network in the remaining chapters. The next sections give an in-depth study of the three basic network structures: the static broadcast networks, wavelength routed networks, and the electronic/optical logically routed networks, covering the characteristics of the optical networks in the access, metropolitan area, and long-haul reach. It discusses the networking picture; network control and management, impairment management and survivability. The last section of the book covers the upcoming technologies of flex-grid and software defined optical networking. * Provides concise, updated, and comprehensive coverage of WDM optical networks * Features numerous examples and exercise problems for the student to practice * Covers, in detail, important topics, such as, access, local area, metropolitan, wide area all-optical and elastic networks * Includes protocols, design, and analysis along with the control and management of the networks * Offers exclusive chapters on advance topics to cover the present and future technological trends, such as, software defined optical networking and the flexible grid optical networks Optical WDM Networks: From Static to Elastic Networks is an excellent book for under and post graduate students in electrical/communication engineering. It will also be very useful to practicing professionals in communications, networking, and optical systems.

Leseprobe

Optical WDM Networks: From Static to Elastic Networks

Devi Chadha

Indian Institute of Technology, New Delhi, India

Copyright

This edition first published 2019

© 2019 John Wiley & Sons Ltd

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.

The right of Devi Chadha to be identified as the author of this work has been asserted in accordance with law.

Registered Offices

John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USA

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

Editorial Office

The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK

For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.

Wiley also publishes its books in a variety of electronic formats and by print‐on‐demand. Some content that appears in standard print versions of this book may not be available in other formats.

Limit of Liability/Disclaimer of Warranty

In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.

Library of Congress Cataloging‐in‐Publication Data

Name: Chadha, Devi, author.Title: Optical WDM networks : From Static to Elastic Networks / Devi Chadha, Indian Institute of Technology, New Delhi, India.Description: Hoboken, NJ, USA : Wiley, [2019] | Includes bibliographical references and index. | Identifiers: LCCN 2018061450 (print) | LCCN 2019000068 (ebook) | ISBN9781119393375 (AdobePDF) | ISBN 9781119393344 (ePub) | ISBN 9781119393269(hardcover)Subjects: LCSH: Wavelength division multiplexing.Classification: LCC TK5103.592.W38 (ebook) | LCC TK5103.592.W38 C53 2019 (print) | DDC 621.39/81–dc23LC record available at https://lccn.loc.gov/2018061450

Cover Design: Wiley

Cover Image: © Wenjie Dong/Getty Images

Preface

The last decade has seen significantly increased use of high bandwidth emerging dynamic applications, such as real‐time multimedia streaming, cloud computing, data center networking, etc. These rapid advances require the next‐generation optical communication networks to adapt to these changes by becoming more agile and programmable, and to meet the demands of high bandwidth and flexibility with much higher efficiency and reduced cost. It therefore becomes necessary to bring out a comprehensive and up‐to‐date book of optical networks. This was the driver for this text.

Writing a book on optical networking entails covering materials that span several disciplines, ranging from physics to electrical engineering to computer science and operations research. The treatment of the material requires uncovering the unique strengths and limitations of the appropriate technologies, and then determining how those are exploited in pragmatic network architectures, while compensating for the limitations. The paradigm shift in optical networking which we are seeing with software defined networking requires clear basic concepts to conduct research and development in these newer optical technologies. It is difficult to develop newer sophisticated technology for different applications without understanding its evolutionary process. Thus, the task of writing such a book becomes quite challenging.

Overview of the Book

This book attempts to cover components and networking issues related to second‐generation optical networks. The second generation of fiber optic networks exploits the capacity of fiber to achieve hundreds of gigabits per second to terabits per second with dense wavelength division multiplexing (DWDM) by using routing and switching of signals in the optical domain. There is now a matured large bandwidth underlying optical technology available with new tools for network control and management. There is also a recognition of the latest directions of optical network deployment and research. These new directions include cost‐effective network architectures tailored to the strengths of current optical transmission and switching equipment, passive optical networks to bring high‐speed access to the end user, hybrid optical/electronic architectures supporting the merging of multi‐wavelength and Internet technologies, and networks of the future based on all‐optical packet switching and software programming. Keeping this view in mind, the book covers the fiber optic wavelength division multiplexing (WDM) networks in ten chapters as detailed below.

Chapter 1 offers an introduction to optical networks with an overview of the fundamentals of network architecture and services provided by it. Chapter 2 gives an overview of the different components needed to build a network, such as transmitters, receivers, amplifiers, multiplexers, switches, optical cross‐connects, etc. The next couple of chapters then focus on the different types of networks. The broadcast and select basic static multipoint networks are given in Chapter 3. Chapter 4 describes passive optical network solutions for fiber‐to‐the‐x access network applications, while Chapter 5 covers the metropolitan area networks, basically the ring structures. Chapter 6 describes the wavelength routed wide area networks and how to overlay virtual networks, for example IP or OTN networks over an underlying second‐generation optical network. Chapter 7 covers the control plane architecture as it has developed through the recent activities of several standards organizations describing the latest developments in optical network control. It gives detailed discussion of generalized multiprotocol label switching (GMPLS) as it applies to optical networks. Going ahead, Chapters 8–10 are devoted to the advanced techniques used for the design of the upcoming technologies for bringing the expanding capability of the present‐day requirements. Chapter 8 covers the effects of signal impairments, survivability, protection, and restoration in optical networks, consistent with the growing importance of optical layer fault management in current networks. As there have been fundamental changes in many aspects of optical networking, Chapters 9 and 10 cover the upcoming technologies of flex‐grid and software‐defined optical networking.

Exercises are provided for most of the chapters, and many of them suggest avenues for future study. The book is meant to offer several different alternatives for study depending on the interest of the reader, be it understanding the current state of the field, acquiring the analytical tools for network performance evaluation, optimization, and design, or performing research on next‐generation networks.

This book has been written primarily as a graduate‐level textbook in the field of optical fiber networks. For this reason, the emphasis is on concepts and methodologies that will stand the test of time. Along with this, the advances in the technology are also discussed throughout the text. An attempt is made to include as much recent material as possible so that students are exposed to the many advances in this exciting field. The book can also serve as a reference text for researchers and industry practitioners engaged in the field of optical networks because of the exposure to much of the advancement in the emerging area. Exhaustive reference lists at the end of chapters are provided for finding any further details which could not be included in the book. The listing of recent research papers should be useful for researchers using this book as a reference.

Acknowledgments

A large number of people have contributed to this book either directly or indirectly, and hence it is impossible to mention all of them by name. First, I thank my graduate students who took my course on optical networks year on year and helped improve my class notes through their questions and comments. Much of the book's material is based on research that I have conducted over the years with my graduate students. I would also like to thank students in my classes for developing many of the figures: Gaurav, Devendra, Nitish, Sridhar, Vishwaraj, Shantanu, and a few others; their efforts are highly appreciated.

I am grateful to my Institute, the Indian Institute of Technology, Delhi, India, for providing a cordial and productive atmosphere, and to my colleagues for many useful discussions during the course of writing the book.

This book could not have been published without the help of many people at Wiley International; in particular, Anita Yadav, acquisitions editor, for taking me through the entire process from start to finish, and Steven Fassioms, my project editor, for orchestrating the production of the book.

On the family front, I'd like to acknowledge the invaluable support given by my dear husband, Dev, during this endeavor, and my loving children Manu, Rati, Rashi, and Varun for their understanding when I needed to spend many hours on the book instead of spending time with them. Finally, I thank the Almighty for giving me the strength to embark upon this project and with His blessing conclude it satisfactorily.

1Introduction to Optical Networks

1.1 Introduction

Any technological development is always driven by the need and demand of the changes in society. The rapid evolution of communication networks from the basic telephone network to the present high‐speed large area networks has come with the social need of people to communicate among themselves, with the increasing user demands for new applications, as well as advances in enabling technologies. The fast changes in the present‐day telecommunication networks are also driven by the user's need to remain connected anytime and all the time, anywhere and everywhere in the world. The new applications, i.e. multimedia services, video‐conferencing, interactive gaming, Internet services, and the World Wide Web, all demand very large bandwidths. Besides this, the user wants the unifying network underneath to be reliable, give the best services, and be cost‐effective as well.

What we need today, therefore, is a communication network with high capacity and low cost, that is fast, reliable, and able to provide a wide variety of services from dedicated to best‐effort services. The available transmission media best suited to meet most of these requirements is optical fiber. Besides having enormous bandwidth in terahertz (~1012 Hz), optical fiber has low loss and cost. It is lightweight, with strength and flexibility, and is immune to electromagnetic interference and noise. It is secure and has many more characteristics to make it an ideal high‐speed transmission line, therefore optical fiber is most suitable to meet the traffic requirements in today's communication network. The enormous quantity of optical fiber laid throughout the world by the end of twentieth century is the foundation of the information super highway of optical network with huge bandwidth today.

1.1.1 Trends in Optical Networking

In order to increase the capacity of point‐to‐point links, optical fibers first replaced the coaxial and two‐wire transmission lines in the existing communication networks. The capacity of the optical links in the network was further increased by using wavelength division multiplexing (WDM) in the laid fiber, thus having several wavelength channels carrying multiple data streams in a single fiber. But all the switching and routing operation in the network still remained in the electronic domain. In the real sense we do not define this network as an all‐optical network. An all‐optical network is the high‐capacity telecommunication network which uses optical technology and components not only to provide large capacity optical fiber links for information transmission but also to do all the networking operations, such as switching and routing (i.e. facilitation of the correct and suitable path) of the signals on the required path, grooming of the low bit rate traffic signals to higher bit rate for better utilization of the enormous capacity of the fiber, and control and restoration functions in the network at the optical level in case of any failure in the network. The operation of the network at the optical or wavelength granularity level has many advantages. As an example, when a single wavelength carries a large number of independent connections and a failure occurs in a fiber cable, it is operationally much simpler to restore services by routing and processing an individual wavelength than to reroute each connection individually. Besides, optical switching functions consume much less power and have lower heat dissipation and footprint compared with their electronic counterparts. With the present optical technology, it is still not possible to achieve all these functions cost‐effectively with ease. Hence, both optical and electronic devices are used, which makes the optical network not purely optical. These networks are indeed hybrid in nature at present, using both optical and electronic technology.

The communication network, which once supported telephone voice traffic only, now carries more data traffic supporting high‐speed multimedia services. At the physical infrastructure level, the available optical components in the optical network can now support traffic at multiple speeds ranging up to tera‐bits/sec (Tbps), with each fiber carrying a large number of wavelengths in the WDM systems. Also, along with the optical component infrastructure, the networks are becoming more flexible and agile due to the adoption of intelligent algorithms and protocols for networking, and therefore the network can now respond to new applications and demands with ease. The trend in optical networks is now toward SDON (software defined optical networking) to facilitate programmability of network operations to further increase agility and to provide users with more control over networking functions, thereby resulting in flexibility in the deployment of new services and protocols, better network utilization, QoS (quality of service), higher revenue generation from the flexibility added in the network, and the user managing the network according to his/her requirements. There is a paradigm shift in optical networking now with SDON, which is a fast‐evolving technology [1–4].

In the rest of the chapter we will give a brief overview of the optical networks and an introduction to the technologies, terminologies, and parameters involved. In order to have some understanding of the functionalities of different types of networks used, somewhat detailed stratification of a generic network is discussed, hoping for an easier understanding of the networks in subsequent chapters where these will be discussed in detail.

1.1.2 Classification of Optical Networks

On the basis of geographical reach, the telecom network can be subdivided in three categories: the access and local area network (LAN), the regional metropolitan area networks (MANs), and the backbone wide area network (WAN). This is shown in Figure 1.1.

Figure 1.1 Classification of the communication network on the basis of reach.

Any two distant users in the access network communicate with each other through the switch/router (or exchange) to the rest of the network infrastructure. The access network can have a span of up to 20 km or so. The access network needs to be cost‐effective as the cost has to be shared among a smaller number of individual users who are connected through it to the shared communication network underneath. The present‐day high‐speed optical solution to the access part is with passive optical networks (PONs), with many variants of the Ethernet passive optical network (EPON), gigabit passive optical network (GPON), and WDM PON, etc. The other lower speed access networks are provided by wireless (WiMax, Wi‐Fi), cable modems, digital subscriber lines (DSLs), higher speed lines (T1/E1), etc. We discuss the access networks in detail in Chapter 4.

The metro network covers distances of a few kilometers to hundreds of kilometers. The metro network aggregates the traffic from the access networks. The technologies used in metro networks are a lot different from those in access networks. Metro networks generally have a ring physical architecture using the legacy Synchronous Optical NETworking (SONET)/synchronous digital hierarchy (SDH), asynchronous transfer mode (ATM), or optical transport network (OTN) networks. A node (router) of the ring or mesh collects or distributes the traffic of the access network connected to the ring while the hub node of the ring on the other end is connected to the edge node of the WAN, as shown in Figure 1.1. The design and management of metro networks are geared toward the different types of traffic streams and services streaming from the access network, and hence the topology can keep changing accordingly. We discuss this in Chapter 5.

The backbone segment of the network, which is invariably a mesh physical topology, spans from hundreds to thousands of kilometers, giving nationwide to global coverage, thus connecting the user in the access of LAN to another user or LAN in some other city, country, or continent. Backbone transport networks are shared among millions of users connected through the multiple metro‐area networks. The long‐haul core networks tend to be more stable in topology, unlike their metro counterpart. In core networks the emphasis is on high‐bandwidth pipes and very high bit rate nodes so that more and more data are moved at high speed and efficiently. Traffic in the core network is transported between routers through high‐speed gigabit links. The backbone routers use IP‐over‐SONET or IP‐over‐ATM‐over‐SONET technology to route IP (Internet Protocol) or the synchronous stream traffic over the wavelength channel fat pipes. Also, if the end user has large data, it can access the backbone network with a direct WDM channel. This is optics in the access with the access data over fiber. Chapter 6 is dedicated to the details of the optical wide area transport networks.

Optical networks could be classified in two generations: first and second. First‐generation optical networks use optical fibers with WDM channels as a replacement for copper cables to increase transmission capacity. However, fiber deployment is mainly for point‐to‐point transmission. All the routing and switching functions are performed electronically at each node. Optical signals go through the optical to electrical conversion first for performing the switching and processing operation in the electronic domain. Later these electronic signals are reconverted to the optical domain for further transmission at each intermediate node as they propagate along an end‐to‐end path from one node to another. Figure 1.2a shows one such connection from source node “D” to destination node “B,” where the optical signal has to go through the O‐E‐O conversion at node 1, node 2, and node 3 in the transponders at each connection to the electronic node. The network nodes reach the electronic bottleneck as they are not able to process all the traffic carried by the node, which is in Tbps in WDM systems because of the limitation of the electronic processing speed. The node traffic includes not only the traffic which is intended for it but also the passing‐by transit traffic intended for other destination nodes. Examples of first‐generation network nodes are the SONET/SDH electronic switch, ATM switch, fiber data distribution interface (FDDI), etc. interconnected with fiber links for transmission. Figure 1.2b shows the schematic of a three‐degree first‐generation network node which is used in the network where the signal transmission is in the optical domain but at each node the optical signal goes through the optical‐to‐electronic and then again electronic‐to‐optical conversion before transmission to the next node.

Figure 1.2 (a) First‐generation optical network. (b) First‐generation 3 degree node.

To overcome the electronic bottleneck for high data speed, the second‐generation optical networks have optical switching nodes instead of electronic switching nodes. These nodes have optical routing and switching functions and are capable of bypassing those signals that carry traffic not intended for the node directly in the optical domain. Hence, they are also called bypass nodes. With the availability of optical devices, such as optical add‐drop multiplexers (OADMs) and optical cross‐connects (OXCs), this has become possible. Figure 1.3 shows the optical bypass capability of the optical node. In this the traffic on the two input ports connected to the node can be bypassed without any O‐E‐O conversion to the two output ports. Any two nodes in this network can be directly connected together by a wavelength (lightpath) in a single‐hop even if they are not connected directly with fiber. WDM optical technology is now widely deployed not only in WANs but also in MANs and LANs.

Figure 1.3 Optical bypass second‐generation network node with add‐drop ports.

Two nodes, if not connected directly by a lightpath, can communicate also using a multi‐hop approach, i.e. by using electronic switching at the intermediate nodes. This electronic switching can be provided by the electronic IP routers, or SONET equipment connected to the optical nodes, leading to an IP‐over‐WDM or a SONET‐over‐WDM network, respectively, in the hybrid network. These ports are also shown in Figure 1.3 as the add/drop ports.

1.2 Optical Networks: A Brief Picture

In the previous sections the optical networks were introduced. Next, we give a brief overview of the technologies, parameters, and some other details in order to understand them better in the subsequent chapters where they will be used and discussed in detail.

1.2.1 Multiplexing in Optical Networks

To use the large bandwidth of the fiber efficiently, signals of multiple connections have to be simultaneously transmitted over the fiber; in other words, multiplexing of signals has to be carried out without overlapping in the spectrum. This is employed for all multi‐user communication in a multi‐access environment. With the present technology, besides the time division multiplexing (TDM) in the electronic domain (Figure 1.4a), WDM is used in the optical domain on the existing fibers, as shown in Figure 1.4b. With WDM technology, multiple optical signals can be transmitted simultaneously and independently on orthogonal optical wavelength channels over a single fiber. Each of the channels has a rate of many gigabits/sec (Gbps), which significantly increases the usable bandwidth of an optical fiber. WDM systems with 88 WDM channels and bandwidth of 100 GHz per channel are now available. Thus, on a single fiber with both TDM and WDM schemes, a total traffic of M(N × b) bits/sec can be carried by wavelength multiplexing M wavelengths, with each wavelength time‐multiplexing N data stream each of b bits/sec.

Figure 1.4 (a) Time division multiplexing. (b) Wavelength division multiplexing.

In addition to the increased usable bandwidth of an optical fiber, we exploit WDM for wavelength routing and switching and restoration in the optical domain. The other advantages of WDM are its reduced processing cost compared with electronic processing, data transparency, and efficient failure handling, as mentioned earlier. As a result, WDM has become a technology of choice in the optical networks. Further, within each WDM channel, it is possible to have frequency division multiplexing (FDM) at the electronic level. In this case, the WDM channel bandwidth is subdivided into many radio frequency (RF) channels, each at a different frequency. Each radio frequency channel carries a different information signal. These RF channels are multiplexed to form a composite FDM signal and then the multiplexed signal intensity modulates the optical carrier. This is called subcarrier multiplexing. Optical code division multiplexing (OCDM) with code division multiplexing access (CDMA) is another optical multiplexing technique used in the networks. Optical time division multiplexing (OTDM) is also one of the possibilities for multi‐user or multi‐access technique to enhance the number of connections in the network, and is being actively researched to make it commercially viable.

1.2.2 Services Supported by Optical Networks

Optical networks are required to support services which are diverse in geographical reach, applications, performance, and characteristics. The network may require connectivity in LANs, MANs, or WANs having a global reach. The optical network supports voice, multimedia, and data services. The services offered may vary in performance, with not so demanding email services (best effort) to high‐capacity, secure, fast services of VoIP (voice over IP), IPTV (Internet Protocol television), telemedicine, or other highly secure financial services requiring very high performance in terms of having minimum or no errors and delay, and with high reliability of the network. All these different services have different requirements. But all the services are to be handled on a single underlying optical network by adaptively handling these with each special requirement for a large user population.

To handle the variety of traffic demands in a large network, the network functions can be broadly divided into three planes:

Transport plane

. The transport plane or data plane has the entire infrastructure, including the electronic switches and routers, the photonic transponders and OXCs, fiber links, etc. associated with the information/data transfer.

Control plane

. The control plane takes care of all the activities concerned with the connection provisioning, reconfiguration, performance, and fault management in the network.

Management plane

. This is the overarching plane which deals with the many and all operations of administration, maintenance, performance monitoring, fault diagnosis, statistics gathering, etc.

The network with the three functional planes is shown in Figure 1.5. We will be discussing the details of these planes in Chapter 7.

Figure 1.5 Functional architecture of optical network.

1.2.3 WDM Optical Network Architectures

There are broadly three classes of WDM optical network architectures: broadcast‐and‐select (B&S) networks, wavelength routed networks (WRNs), and linear lightwave networks (LLNs).

1.2.3.1 Broadcast‐and‐Select Networks

A generic WDM broadcast‐and‐select network consists of a passive optical star coupler network node connecting the user nodes in the network. Each of these user nodes is equipped with one or more fixed or tunable optical transmitters and one or more fixed or tunable optical receivers. Different nodes can transmit messages on different wavelengths simultaneously. The star coupler combines all these messages and then broadcasts the combined message to all the nodes. A user node selects a designated wavelength to receive the desired message by tuning its receiver to that wavelength.

The single‐hop B&S networks are all‐optical; a message from the network user node once transmitted as light reaches its final destination directly, without being converted to electronic form in between. In order to support high data transmission in these networks we need to have optical transmitters and receivers at the nodes that can tune rapidly. The main networking challenge in these networks is the coordination of transmissions between the various user nodes connected to the central passive optical star coupler. In the absence of coordination or efficient medium access control protocol, collisions occur when two or more nodes transmit/receive on the same wavelength at the same time. To support fast switching efficiently in B&S networks, a multi‐hop approach can be used, which avoids rapid tuning.

The advantage of B&S networks is in their simplicity and natural multicasting capability (ability to transmit a message to multiple destinations). However, they have limitations. First, they require a large number of wavelengths, typically at least as many as there are user network nodes in the network, because there is no wavelength reuse possible in the network. Thus, the networks are not scalable beyond the number of supported wavelengths. Second, they cannot span longer distances since the transmitted power is split among various nodes and each node receives only a small fraction of the transmitted power, which becomes smaller as the number of nodes increases. For these reasons, the main application for B&S is for high‐speed local area and access networks. The other topologies which are used in these networks are the folded bus and the tree topology. We will be discussing the single‐hop B&S networks in Chapter 3.

1.2.3.2 Wavelength Routed Networks

Wavelength routed WDM networks have the potential to avoid the three problems in the broadcast networks: the lack of wavelength reuse, power splitting loss, and scalability to WANs. A WRN consists of routing (or bypass) nodes interconnected by fiber links in an arbitrary mesh topology. The end user node is connected to a network node via a fiber link. Each end node is equipped with a set of optical transmitters and receivers for sending data into the network and receiving data from the network, respectively, both of which may be wavelength‐tunable. In a WRN, a message can be sent from the source node to the destination node using a wavelength continuous route called a lightpath on one of the WDM channels, without requiring any O‐E‐O conversion and buffering at the intermediate nodes. This process is known as wavelength routing. The destination node of the lightpath accesses the network using receivers that are tuned to the wavelength on which the lightpath operates. A lightpath is an all‐optical communication path between two nodes, established by allocating the same wavelength throughout the route of the transmitted data. Thus, it is a high‐bandwidth pipe carrying data up to several gigabits per second, and is uniquely identified by a physical path and a wavelength. The requirement that the same wavelength must be used on all the links along the selected route is known as the wavelength continuity constraint. Two lightpaths cannot be assigned the same wavelength on any single fiber. This requirement is known as distinct wavelength assignment constraint. However, two lightpaths can reuse the same wavelength if they use disjoint sets of links. This property is known as wavelength reuse. A WRN is illustrated in Figure 1.6.

Figure 1.6 All‐optical wavelength routing network architecture.

The simultaneous transmission of messages on the same wavelength over fiber link disjoint paths or wavelength reuse property in WRNs makes them more scalable than B&S networks. Another important characteristic which enables WRNs to span long distances is that the transmitted power invested in the lightpath is not split and is sent to the relevant destination only. Given a WDM network, the problem of routing and assigning wavelengths to lightpaths is of paramount importance in these networks. Good algorithms are needed in order to ensure that functions of routing and wavelength assignment are performed using a minimum number of wavelengths. Connections in WRNs can be supported by using either a single‐hop or a multi‐hop approach. In the multi‐hop approach, a packet from one node may have to be routed through some intermediate nodes with an O‐E‐O conversion before reaching its final destination. At these intermediate nodes, the packet is converted to electronic form and retransmitted on another wavelength. These attractive features – wavelength reuse, protocol transparency, and reliability – make WRNs suitable for WANs.

1.2.3.3 Linear Lightwave Networks

In the case of WRNs we multiplex several wavelengths and each of these orthogonal wavelengths can form a lightpath to route a connection in the network. In LLNs, meanwhile, we use waveband, which is a collection of a number of orthogonal wavelengths to route over the network as a unit. In these networks several wavebands are multiplexed on a fiber and several wavelengths are multiplexed in each waveband. The routing nodes de‐multiplex and multiplex wavebands but not wavelengths within a waveband. Like the wavelength, the waveband‐routed networks have the property that each waveband channel can be recognized in the optical network nodes (ONNs) and routed individually and hence network capacity can be improved with the spectrum reuse. Since an LLN node does not distinguish between wavelengths within a waveband, individual wavelengths within a waveband are separated from each other at the end node's optical receiver demultiplexer. The two constraints of waveband continuity and distinct waveband assignment as in WRN apply to LLNs. Further, there are two routing constraints unique to LLNs: inseparability, that is, channels belonging to the same waveband when combined on a single fiber cannot be separated within the network; and distinct source combining, that is, on any fiber, only signals from distinct sources may be combined. We will be limiting our discussion to wavelength routing networks and not considering the LLNs any further.

1.2.4 Services Types

The optical networks are used for different applications, varying from voice and data to other multimedia transmission. In general, network services for all the applications provided can be classified as follows:

Connection‐oriented services

. These services are possible on circuit‐switched networks, where the connection over the network is established before the traffic can be sent and is maintained until the connection is dropped. Examples of such services/protocols are the legacy telephone services, ATM,

TCP

(

transmission control protocol

),

MPLS

(

multiple protocol label switching

) services. All nodes have to maintain state information once the handshake procedure has been executed. Thus, QoS is maintained as decided by the

service level agreement

(

SLA

).

The connection‐oriented services can have two types of connections:

Dedicated connections

. In the circuit‐switched networks where the path is dedicated for a connection over long periods of time, possibly for months, days, or hours, it is said to be a dedicated connection. The user has a service agreement with the service provider for the connection. The connection is disrupted only when fault occurs and is then quickly put on another path.

Demand‐oriented connections

. The demand‐oriented connections are also circuit switched and provided to the user when demanded. As the network resources are shared, the services are provided on available free resources, otherwise the connection is blocked or dropped. Once the demand is accepted, the path is reserved and kept connected as demanded by the client. The demand may be over several seconds or milliseconds.

Connectionless services

. Connectionless service does not require the establishment of any connection prior to sending data. The sender starts transmitting data to the destination according to the access control in use. Therefore, the connectionless services are less reliable than connection‐oriented services. These are packet switched services; the connection is made as the data flows over the network, with IP providing the best effort (BE) services. They are much more economical but with reduced QoS compared with connection‐oriented services.

1.2.5 Types of Traffic

Static and dynamic

. The traffic can be either fixed or changing dynamically with time, both in its source destination pair connection and in terms of demand size. In the case of static traffic demand, connection requests are known beforehand or a‐priori with the estimation of long‐term traffic demands of the network. This is expressed as the traffic matrix, which is static. In the case of dynamic traffic, the connection requests change with time in the network in random fashion and so does the traffic matrix.

Stream or synchronous traffic

. The stream traffic flow is uniform and constant and with the same clock period for every switch in the network – for example, SONET/SDH with fixed rate.

Asynchronous or random traffic

. The random traffic is dynamic with random arrival rate and length. It can take some probability distribution, i.e. uniform, Poisson, etc., or it may be bursty in nature, for example Internet traffic.

1.2.6 Switching Granularities

In optical networks we talk of different types of granularities, such as granularity of traffic, of switching speed, or of connection capacity.

Depending on the traffic demands of the various clients, optical networks need to be able to provide connections of different traffic demand granularity, such as 1 Gbps, 40 Gbps, etc. Also, the granularity of the connection's switching speed can be either fast or slow according to the required application and performance, QoS, etc. In the optical switching network, the switching can also be on the basis of granularity of connection capacity, such as whether it is space switching from one fiber to another, waveband switching, wavelength switching, or time‐slot switching. The connection capacity in space switching is of the order of Tbps, i.e. the total capacity of all the WDM channels in a fiber, while in the case of time‐slot switching it is a single user connection, so the order of Mbps may be involved. The connections provided by optical switching networks must be able to adapt to time‐varying network conditions and traffic requirements dynamically in order to optimize network performance and utilization of network resources.

1.2.6.1 Optical Circuit Switching

For the connection‐oriented services in circuit switching we can perform switching at the following granularity:

Space or fiber switching

. Any connection from one incoming fiber can be transferred to another outgoing fiber by switching in space. This is the coarsest level of switching in terms of granularity, as a large number of connections carrying a huge amount of data are switched in this case.

Waveband switching

. This is the next level of lower switching in which a waveband is switched at the node. A waveband consists of a number of wavelengths. These wavebands are switched independently from each other – that is, network nodes are able to switch individual wavebands arriving on the same incoming fiber to a different waveband in the outgoing fiber.

Wavelength switching

. This is a further finer granularity in which individual wavelengths in a waveband or a fiber itself are first de‐multiplexed to the level of individual wavelengths, and then any of these wavelengths can be independently switched in the input fiber to the other wavelength in the output fiber.

Sub‐wavelength

. Sub‐wavelength switching is also possible when OTDM is possible. Each slot in the TDM time frame carries single client data which can be switched to another slot. But this is still at the research stage. TDM in the electronic domain is done to multiplex lower rate data connections to accommodate low rate client data, as in SONET/SDH systems.

1.2.6.2 Packet Switching for Bursty Traffic

Circuit switching in optical networks is not economical in case of random traffic load. If the random bursty traffic is provided to a dedicated connection, then large bandwidth is wasted when low traffic density is greater than its statistical average. Also, on the contrary, the performance will deteriorate considerably when optical burst with high packet density arrives. In such situations switching is recommended for the random traffic by individual packet as it arrives at the switch. Packet switching is done at two granularity levels:

Individual packet switching

. In the case of packet switching, each packet has a header with source‐destination address which is recognized at the nodes. After reading the address, a node forwards the packet to the next node on a wavelength channel. Thus, individual packets are switched independently from one to another node.

Burst switching

. Unlike in the case of packet switching, packets are aggregated at the ingress node and sent as bursts across the network. For each burst a reservation control signal is sent on a dedicated control wavelength channel prior to sending the burst on one of the data wavelength channels after a pre‐specified offset time.

1.3 Optical Network Layered Architecture

Until the early twentieth century before the fiber was well laid, only limited data traffic was carried over the existing telephone network which used to carry voice services. With optical fiber now the main transmission medium, the networks have become more data‐centric rather than voice‐centric, and digital transmission standards of higher bit rate, such as SDH/SONET, ATM, and IP networks, are well developed for this purpose. Internet Protocol is being used for transmitting all types of traffic, be it voice, video, or data, with all types of multimedia services on the optical fiber network. In other words, the services supported by the underlying optical network now are vast, extremely diverse in terms of connectivity, bandwidth, performance, reach, cost, and with many other features. Therefore, the network has to adapt to different special features for each type of service.

With the ubiquitous optical network to provide the variety of services mentioned above, the complexity of the network increases. Different components in the network have to perform a variety of functions. Therefore, as the OSI (open systems interconnection) model in computer networks, we can conceptually divide the functionalities of any node in the network in the constituent layers with client‐server relations with the neighboring layers. The layered architecture allows flexibility not only in designing the equipment but also in design and implementation of the network. By employing a layered model, designers are left with enough leverage to add functions at each layer and to concentrate only on certain functions in the network provided by the specific layer while maintaining interoperability with other layers. The layered model helps in understanding and designing the optical network architecture, which is flexible, robust, and scalable.

In an optical network, when a user makes a request for connection, the source node makes a connection to its destination end by passing through a number of intermediate network elements or nodes. A variety of functions has to be carried out at each network element, such as framing of the packets, grooming, multiplexing, routing, and switching. In the layered architecture each layer carries out specific functions as required by the layer above it, then the layer after performing the required functions, instructs the layer below to perform the necessary specified function. Thus the data flows between each layer. Each intermediate network element, between the source and the destination node along the path, has a certain number of layers depending on the network element's function, starting from the lowest layer up to a certain higher layer in the hierarchy.

In the seven‐layered ISO (International Organization for Standardization) structure there are two types of layers: host and media. The upper four layers – the transport, session, presentation, and applications layers – are the host layers and the lower media layer has three sub‐layers. These sub‐layers are concerned with the functionalities of the data flow over the common network media. The lowest layer in the hierarchy is the physical layer, followed by the data link layer and the network layer. The physical layer provides the pipe with a certain bandwidth to the above layer. The pipe can be an optical fiber, a coaxial line, or any physical channel. The data link layer provides multiplexing, de‐multiplexing of channels, framing of the data to be sent over the physical layer, etc. The network layer above the data link layer provides the end‐to‐end service to the message. In the present text we will be concentrating on these lowermost media layers in the optical networks.

The functional layered structure of a multilayered WDM optical network is shown in Figure 1.7. In the case of optical networks we have three layers for the functional layer abstraction: client or user network, logical, and physical. Each adjacent layer has a client server relationship with the lower layer serving the adjoining upper layer. The bottom layer, the physical layer, serves the adjoining second layer by providing the required basic optical channel bandwidth service to logical network nodes in the logical layer. As the physical layer is a fiber pipe, it therefore encompasses optical multiplexing, transport, and switching in the optical devices based on optical technology. The logical switching nodes (LSNs) of the logical layer, including IP routers, Ethernet switches, ATM switches, SONET/SDH switches, and OTN switches, are all based on electronic technology. The protocols associated with each of these switches do the formatting and partitioning of the data and move the data from source to destination over the optical physical layer by specific methods. Each of these LSNs organizes the raw offered capacity by the physical layer to the needs of the clients in the user layer, which is also called the service or application layer. The application layer includes all types of services, such as voice, video, and data. Thus, the several logical networks in the logical layer can provide a variety of specialized services to a connection demanded by the upper client layer.

Figure 1.7 The multilayer optical network model.

For example, the SONET switches in the logical layer will electronically multiplex the low‐speed data streams of the client layer into higher‐speed streams, which in turn use optical wavelength channels provided by the physical layer to transmit the multiplexed data as an optical signal. The SONET channels can support a wide variety of services, such as telephone services, data, video, etc. In another case Internet services can be provided to some users through the IP layer sitting on the ATM over the SONET when reliability of ATM services is required before multiplexing them in the SONET layer. For certain IP we may have the IP layer directly over the optical physical layer, and yet in another case when the application layer connection requires a very high bandwidth, a raw wavelength service may directly be over the physical layer bypassing the logical layer, thus providing totally transparent optical network functions. Finally, in the physical layer, multiple wavelengths or lightpaths are optically multiplexed, switched, and routed in the WDM fiber physical layer, providing large bandwidth connectivity to the upper layers.

Next, we first give a simplistic view of how connection is made in an optical network before explaining in detail the functionalities of each layer. A typical optical network picture is given in Figure 1.8. The ONNs, which are OXC or wavelength or space switches, are interconnected with fibers in the physical optical layer. These ONNs are responsible for connecting different fibers as per the requirements of the logical connection demanded by the LSNs. These optical nodes are connected to the LSNs through the optical line terminal (OLT) or the network access terminal (NAT