Telektronikk 2/3.2001

1 Introduction

Telecommunications as we have known it, is

being literally transformed with the convergence

of voice services with data and multimedia ser-

vices, mobility requirements, and the presence

of several competing players in the market. The

industrial age is allegedly over, the world is

going digital, global and on the net, and we have

entered the information era where new para-

digms are dominating corporate, social as well

as private life. Despite some ups and downs in

the forecasts, the fact remains that data traffic

has increased dramatically in the past years and

has now surpassed voice traffic in volume, signi-

fying a soft transition to the new economy. But

how do our networks catch up with this (r)evolu-

tion?

Today’s SONET/SDH network infrastructure

provides a guaranteed level of performance and

reliability for voice calls and leased lines. Exist-

ing networks have been designed for telephony

and are thus adequate for handling static traffic

patterns but rather inefficient in handling the

new traffic patterns that are dominated by data

traffic. In contrast to traffic generated by tele-

phony, data traffic patterns are more unpre-

dictable, asymmetric in terms of load distribu-

tion, and bursty in nature. Convergence in the

applications front, the potential for more sophis-

ticated services and the requirement for tailored

billing mechanisms in view of the expanding

competition in the liberalised telecom market,

create a positive feedback loop that further

strengthens the requirement for more dynamic

and flexible networks. In addition, because of

the dramatic increase in the capacity carried by

each cable (of the order of Tbit/s), it is manda-

tory to have reliable and fast ways to restore the

network in case of fibre cut or other failure and

to be able to prioritise traffic depending on the

carried service.

All in all, there is clearly a need to realise net-

works that can adapt to changing traffic require-

ments and can make a good use of the network

resources, with fast automatic reconfiguration,

efficient traffic engineering, and service differ-

entiation. These will allow a rationalised – i.e.

economical – use of the network, increase the

synergy between the different network platforms

and enable the introduction of new value-added

services.

At the physical layer, the tremendous increase in

network capacity has been facilitated by techno-

logical advances in optical transmission systems,

i.e. the deployment of dense wavelength division

multiplexing (DWDM). Optical technology has

followed a rather explosive growth curve in the

past five years or so, primarily driven by the

growth in Internet traffic, and – not least – future

growth expectations. Optical transmission is

widely used in most parts of the network world-

wide, namely from transatlantic and pan-Euro-

pean connections, to national connections be-

tween cities as well as within cities. Metropol-

itan area networks are being built and becoming

predominantly fibre-based as many businesses

require high-bandwidth connections, and fibre

installations are taking off substantially also in

the residential customer access area.

The role optical technology has played so far has

been that of a “dumb fat pipe” – as it has been

described with a certain touch of endearment.

The network functionality potential of optical

technology is still largely unexploited and net-

works are today built using layer upon layer with

duplicated functionality. This is rather uneco-

nomical both in terms of cost and in terms of

time consumption, hence there is strong consen-

sus in the engineering community that network

architectures need to be rationalised, piles of

unnecessary equipment removed, and some lay-

ers collapsed. What is not quite clear yet is

exactly how this ought to be realised – what

architecture is most efficient, future proof, and

feasible – as well as what role optical technology

will play to best facilitate network evolution.

Optical Network Functionality: From

“Dumb Fat Pipes” to Bright Networking

EVI ZOUGANELI

Optical technology has experienced an explosive growth in the past years that has enabled multi-terabit

optical transmission over several hundreds of kilometres. Yet the potential of optical networking is far

from being exploited. Optical network functionality may be the answer to efficient and reliable data-cen-

tric networks, a technology that complements IP. This article aims at giving an overview of the driving

forces behind optical networking, the potential offered by it, the challenges encountered, and the current

state-of-the-art.

Evi Zouganeli (38) studied

Applied Physics at the Univ. of

Patras, Greece, and after earn-

ing the national postgraduate

scholarship in 1986, she ob-

tained the MSc in Telecommuni-

cations (1988) and PhD in Opto-

electronics (1992), both from

University College London, UK.

She is currently Senior Research

Scientist at Telenor R&D with

interests focussing on optical

network architectures. She

joined Telenor R&D in 1994 and

has since worked on high

capacity optical transmission,

optical networking and migra-

tion scenarios, as well as strate-

gies for the upgrading of the

Norwegian network – both for

Telenor BUs and in European

collaboration projects. She is

member of a number of inter-

national Technical Committees.

Prior to joining Telenor, she was

with the Federal Institute of

Technology, Zürich, Switzer-

land, where she worked with

optical switching and high

capacity optical networks.

[email protected]

346