You are on page 1of 44

LTE has WiMAX

in its sights
PLUS
M O B I L E B R O A D B A N D S E R V I C E S
Vol. 42 No. 2 March 2008 THE GLOBAL AUTHORI TY FOR THE BUSI NESS OF TELECOM
w
w
w
.
t
e
l
e
c
o
m
m
a
g
a
z
i
n
e
.
c
o
m
Official Provider
Daily Newspaper
Online News Service
3G awaits Olympic start
The new mobile home
Breaking the backhaul bottleneck
VIEW FROM THE TOP
Mark Kaish
VP, Cox Communications
TC_CoverMAR.P1.indd C1 2/29/08 10:25:06 AM
UCan get the facts you need about
Unied Communications.
2008 Avaya Inc. All Rights Reserved. *Standard text messaging rates apply.
NO PURCHASE NECESSARY TO ENTER OR WIN. Sweepstakes begins: 12:01 AM (ET) on 11/01/07; ends: 11:59 PM (ET) on 3/31/08. Open to legal residents of the 50 U.S. &
DC & Canada (except the province of Quebec), age 18 or older who are Avaya customers or potential Avaya customers as of 11/01/07 or individuals involved in IT decision
making. Visit www.avaya.com/accelerate for complete Ofcial Rules. Void in PR, the province of Quebec & where prohibited by law. Sponsor: Avaya Inc.
Get the exclusive Forrester Consulting study, compliments of Avaya.
Overcome communication bottlenecks, boost revenue, and improve project management
processes. Simply download Unied Communications Accelerates Business Processes,
an exclusive Forrester Consulting study. Youll learn how UC can trigger hard cost savings,
improve business agility, and make your employees more productive. Plus, youll get
compelling facts that can help you build a convincing business case.
Text AVAYA to 49737 to learn how to win an Archos WiFi
Personal Media Player (a $250 value) and receive the
Forrester Consulting study,* or go to avaya.com/accelerate
C2Avaya/0308.indd C2 2/28/08 2:20:49 PM
Vol. 42 No. 2
March 2008
ON THE COVER
18
LTE has WiMAX
in its sights
The battle for 4G intensifes as
LTE supporters claim growing
momentum
Ken Wieland
FEATURES
24
3G awaits Olympic start
The Chinese governments pledge
to launch 3G in Beijing this summer
puts TD-SCDMA, the countrys
homegrown standard, under the
spotlight. Can it deliver?
Lin Sun
30
The new mobile home
Fixed-mobile convergence has been
a disappointment until now. Could
femtocells be the answer?
Iain Morris
34
Breaking the backhaul
bottleneck
Wireless operators search for new
alternatives
Sean Buckley
36
CDMA hits
the twilight zone
In the face of waning support,
CDMA is now an end-of-life technol-
ogy that will see a long twilight
Phil Marshall
M O B I L E B R O A D B A N D S E R V I C E S
THE GLOBAL AUTHORI TY FOR THE BUSI NESS OF TELECOM
TELECOMMUNICATIONS (USPS 396-270)(ISSN 1936-993X) is published 7 times per year
by Horizon House Publications, Inc., William M. Bazzy, CEO. TELECOMMUNICATIONS is
issued without charge, upon written request, to qualifed persons working directly in
the communications industry, including governmental and educational facilities. OTHER
SUBSCRIPTIONS: domestic: $85.00 per year; foreign: $134 per year; back issues (if available)
and single copies $10 domestic and
$20.00 foreign. Periodicals postage paid at Norwood, MA 02062, and additional mailing
of ce. POSTMASTER: Please send address corrections to TELECOMMUNICATIONS, P.O. Box
3255, Northbrook, IL 60065-3255.
COPYRIGHT 2008 by Horizon House Publications, Inc., Microflm copies of
TELECOMMUNICATIONS are available from University Microflms, 300 N. Zeeb Road, Ann
Arbor, MI 48106, and Bell & Howell, Old Mansfeld Road, Wooster, OH 44691.
Horizon House also publishes Microwave Journal.
Photocopy Rights: Permission to photocopy for internal or personal use or the internal or
personal use of specifc clients is granted by for users through Copyright Clearance Center
(CCC), provided that the base fee of $5.00 per copy of the article, plus $1 per page is paid
directly to the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923, U.S.A.
(508) 750-8400. For Government and/or educational classroom use, the rate is .03 Cents per
page. Please specify: ISSN 0-278-4831.
TELECOMMUNICATIONS can also be purchased on 35mm flm from University Microflms,
Periodic Entry Department, 300 N. Zeeb Road, Ann Arbor, MI 48106; (313) 761-4700.
Reprints: For requests of 500 or more reprints, please contact Gillian Donovan on Tel: +44 20
7596 8744, Fax: +44 20 7596 8749. Claims for missing issues must be fled within 90 days of
date of issue for complimentary replacement.
ADVERTISER INDEX MARCH 2008
Arbor Networks ...............................................................................................7
Avaya Inc. .............................................................................................. Cover 2
Ceragon Networks .............................................................................21, 22-23
CommunicAsia 2008 .......................................................................................8
CTIA Wireless 2008 ................................................................................ Cover 3
International Engineering Consortium ..........................................................27
ITU-Shows ...............................................................................................15, 29
Juniper Networks.............................................................................................5
M2M Zone .....................................................................................................33
NEC Corporation.............................................................................................13
NTT DoCoMo, Inc. ..........................................................................................11
Telecommunications......................................................................................41
Verizon ................................................................................................... Cover 4

www.telecommagazine.com

March 2008 Telecommunications

3

38
Green to go for fber
FTTH Council Europe claims fber is
environmentally friendly
Kendrick Struthers-Watson
DEPARTMENTS
06
Thinking Out Loud
Subprime pricing?

09
News Analysis
Not fazed by Microsoft or Android,
Symbian says ... Telcos regulatory
strategy at risk ... EC throws down
data roaming gauntlet ... Europe
plays FTTH catch-up
14
China Watch
A glimmer of hope for IPTV
16
View from the Top
Mark Kaish, VP,
Cox Communications
40
TechSpeak
IP service control goes beyond
switching & routing
42
Endpoint
M:Metrics highlights the key failings
of mobile TV
We are absolutely full
speed ahead, and I dont
see us backing off at all
Mark Kaish, Cox Communications
24
16
34
MART02-TOC.FINAL.indd 3 3/4/08 1:51:02 PM
telecommagazine.c
m
W E B I N A R S
Listen in as Executive Editor
Sean Buckley Interviews:
GSM Americas, Chris Pearson, President*
CDMA Development Group, Perry LaForge,
Founder, Executive Director and Chairman*
FemtoForum, Simon Saunders, Chair*
WiMAX Forum, Ron Resnick, President*
4

Telecommunications March 2008

www.telecommagazine.com

Telecommunications editors travel to major


industry events to bring you news and
updates from the show foor.
Our editors will be onsite at
CTIA Wireless 2008, bringing
you daily updates straight from the show foor
(Las Vegas, 1-3 April)
Visit the Events Calendar for a listing of all
industry events worldwide.
www.telecommagazine.com/events
Onsite
Take a look at whats online in March:
New case studies on service providers rolling out
FTTH networks
Cable operators tap into the wireless backhaul
opportunity
Sean Buckley interviews key executives at Bell Labs
www.telecommagazine.com/audio
*subject to change
PON and Ethernet in FTTH Networks: Incompatible or
Complementary?
Carrier Ethernet Features for Next Gen Services
Generating Revenue from Addressable IPTV
Advertising
Using Femtocells in Home Networks
Optical Transport Is Going to POT
Presented in partnership with Infonetics Research
www.telecommagazine.com/webinars
Mobile and Wireless
Wireless operators come to a crossroads as they look
at what path to take for their 4G network rollouts. The
question has come down to whether WiMAX or LTE is
the best way.
Next Gen Backhaul: 3G, 4G, and Beyond
Hosted by Infonetics Research
RoomN253
Telecom Live
the latest telecom news at your ngertips
Join us for two Special Interest Seminars at CTIA Wireless 2008:
M2M Carrier Strategies: The New Rules
Hosted by Yankee Group
Room N103
Wednesday, April 2, 9:00 am- 4:45 pm
Visit www.telecommagazine.com/CTIA08
for details
Tuesday, April 1, 12:30-5:00 pm
Visit www.telecommagazine.com/m2mseminar
for details
MART02-TOC.FINAL.indd 4 3/4/08 1:57:54 PM
>> The curse of legacy carrier Ethernet: Increased bandwidth stress, incredible traffc strains,
incomprehensible implementation and operations demands.
So turn to Juniper Networks for innovative plug-and-play technologies that move you
forward, faster increasing scalability, elevating performance and instantly delivering a
high-performance carrier Ethernet network. Our globally trusted JUNOS operating system
eliminates operational complexity, while supporting next-generation services such as IPTV,
VoD and VoIP. So it reduces costs and maximizes revenue opportunities. The industrys
most reliable, cost-effcient MX-series platforms deliver superior capacity, port density and
functionality. Its how Juniper can make any network better: www.juniper.net/carrierethernet


2
0
0
7

J
u
n
i
p
e
r

N
e
t
w
o
r
k
s
,

I
n
c
.
1 . 8 8 8 . J U N I P E R
p5/Juniper/M08.indd 5 2/28/08 2:23:00 PM
Michel Zoghob
Publisher
mzoghob@telecommagazine.com
JoAnne Emery
Associate Publisher
jemery@telecommagazine.com
Jared Bazzy
Website Publisher
EDITORIAL
Ken Wieland
Editor in Chief
kwieland@telecommagazine.com
Sean Buckley
Executive Editor
sbuckley@telecommagazine.com
Jim Barthold
Senior Editor
jbarthold@telecommagazine.com
Georgia Mullen
Managing Editor
Samantha Bookman
Web Editor
Tom Nolle
Contributing Editor
tnolle@cimicorp.com
Michael Kennedy
Contributing Editor
mkennedy@nspllc.com
Lin Sun
Contributing Editor
ADVERTISING
JoAnne Emery
Eastern US/Eastern Canada
jemery@telecommagazine.com
Taryn OFlaherty
Western US/Western Canada
tofaherty@telecommagazine.com
Richard Vaughan
International Sales Manager
rvaughan@horizonhouse.co.uk
OPERATIONS
Edward Kiessling
Trafc Manager
Ken Hernandez
Ad Trafc Assistant
Janice Levenson
Graphic Designer
Robert Bass
Director of Production
& Distribution
Michelle Carletti
Circulation Manager
Eugenie Hardie
Ofce Manager
Lindsay Frankel
Marketing Associate
Greg Lamb
Multimedia Designer
Natalie Zhuravleva
Web Programmer
Michael OBrien
Website Production Designer
The Global Authority for the Business of Telecom
William Bazzy
Chairman of the Board
William M. Bazzy
Chief Executive Ofcer
Ivar Bazzy
President
Jared Bazzy
Vice President
CORPORATE OFFICERS
Headquarters 685 Canton Street Norwood, MA 02062
Editorial and Sales Tel: (781) 769-9750 Fax (781) 762-9071
Circulation tc@omeda.com Tel: (847) 291-5216
Europe 46 Gillingham Street London SW1V 1HH
Editorial Tel:+44-20-7596-8739
Sales Tel:+44 20 7596 8740 Fax:+44 20 7596 8749
Horizon House also publishes Microwave Journal
EDITORIAL OFFICES
T H I N K I N G O U T L O U D
6

Telecommunications March 2008



www.telecommagazine.com


Subprime
pricing?
FLAT-RATE PRICING is great for the customer, provided the rates are low enough of
course. There are no nasty surprises on the bill and its simple to understand. From the carriers
perspective there could also be an upside if fat-rate tarifs can attract a suf cient number of
extra customers to ofset any losses from the presumably lower rates.
But the race for customers through fat-rate pricing is fraught with danger. At press time,
Verizon, AT&T and T-Mobile USA had each announced unlimited nationwide calling plans for
their wireless customers at US$99.99 per month. Fearing an escalating price war, the immediate
reaction of the stock markets was negative and far reaching. Not only did stock prices fall for the
three carriers, but Sprint Nextels share price fell, too, as did Vodafones (which has a 45 percent
stake in Verizon Wireless).
Aside from the worrying prospect of ever-decreasing prices to stay ahead of competitors, fat
rates allow heavy users on the network to be charged the same amount as lighter users (an odd
concept in most retail sectors apart from, perhaps, all-you-can-eat bufets). It also encourages
lighter users to be heavier users. With more and more capacity being soaked up on the network,
and the costs that entail, carriers are in danger of being penned into a corner if the scope for
usage charging is curtailed.
Their position could become even more awkward with ever more competitive fat-rate data
packages. With all-you-can-eat bufets, the thinking must be that owners will be able to sell
more drinks (at much higher margins than the food) to more customers. With fat-rate data
pricing, though, there is a risk carriers give away the equivalent of the higher-margin drinks that
you have to pay for in those all-you-can-eat bufets. In addition to its unlimited nationwide call-
ing plan for US$99.99 per month, T-Mobile USA has already said it will ofer unlimited text and
picture messaging for US$14.99 per month.
The need for carriers to develop services customers are willing to pay for (in addition to fat
rates) is a clear goal, yet fallout from the collapse of the subprime mortgage market in the United
States is sure to make this much harder to achieve. Its reasonable to expect consumer belt-
tightening in the U.S. will continue to put most retail sectors under pressureparticularly those
dealing in luxury goodsbut mobile operators heavily reliant on prepaid subscribers (who gen-
erally have less money to spend) also look vulnerable. Other sources of hitherto strong wireless
revenue growth, such as ring tones and music downloads, may also be adversely afected.
Sprint Nextel, the third largest mobile operator in the U.S., has already apportioned some
of the blame for its recent poor fnancial performances on the subprime mortgage mess. The
economic squeeze has no doubt been a factor in explaining the companys fnancial woes, but
it doesnt tell the whole story. Sprint Nextel has a reputation for bad network quality and lack of
mass-market appeal. This helps explain why Sprint Nextel has reported enormous net losses of
contract customers (more than 600,000 in 4Q07), but AT&T and Verizon have both managed to
make net gains in customer contracts.
For its part, AT&T has tried to downplay the impact a weakening economy will have on its
business. This positioning comes after a sharp dip in AT&Ts share price earlier this year when
CEO Randall Stephenson seemed to indicate things had taken a turn for the worse by pointing
out some customers were not paying bills and had to be disconnected. AT&Ts tack is now to
point out that connectivity is an essential service and so provides some in-built immunity from
economic downturns and lower customer spending.
The major U.S. wireless carriers seem to be entering choppier waters.
Ken Wieland, Editor in Chief
MART03-Commentary.FINAL.indd 6 2/28/08 2:43:06 PM
DETECT
HOW DO YOU PLAN TO
DETECT
NETWORK OPPORTUNITIES BEFORE
YOUR COMPETITORS DO?
By closely monitoring the growth of applications on your network, its easy to
transform new users into an entirely new source of revenue. Can you see it?
Arbor Peakflow

SP turns pervasive network visibility into business intelligence


for a clear competitive advantage. Learn more at www.arbornetworks.com/detect
How networks grow

p07Arbor0308.indd 7 2/28/08 2:24:26 PM


p08CommunicAsia0308.indd 8 3/4/08 1:56:03 PM
Not fazed by Microsoft
or Android, Symbian says
NOKIA-BACKED SYMBIAN,
which supplies mobile OS for smartphones, has
posted a strong set of fnancial results for 2007.
More than 77.3 million Symbian smartphones
were shipped to customers worldwide, a 50
percent increase on the number shipped last
year (52 million). This gave Symbian a year-on-
year turnover boost of 17 percent, taking its
2007 revenue to 194.3m.
Symbian sales during 4Q07, however, fell
short of Nomura Banks expectations: 22.4
million units were shipped, which was 1.6
million below Nomuras forecast that had as-
sumed the normal seasonal pick upduring
4Q07. We believe [the lower than expected
sales] has a lot do with the trend toward
higher specifcation devices and the failure
of the N81 to ship big volumes,says Richard
Windsor, an analyst with Nomura Bank, in a
research note.
Symbian CEO Nigel Cliford anticipates
strong growth coming from the mid-range
market this year, although he does not
provide specifc forecasts. (Symbian quotes
analysts who predict there will be 1 billion
mobile phones based on an advanced open
OS in the market by 2011.)
By the end of 2007, the cumulative
number of Symbian smartphones shipped
reached 188 million since frst coming onto
the market in 2000. Symbians share of the
overall mobile phone market now stands at
around 7 percent.
With three devices being sold every
second during 4Q07 on average, we expect
to pass the 200-million milestone [before the
end of February],Cliford says.
Vendors demonstrated four new mid-
range Symbian smartphones at the Mobile
World Congress (MWC) event held in Barce-
lona last month: two by Sony Ericsson, one
by LG and one by Samsung. There are now
more than 140 Symbian smartphone models
on the market (108 by the end of 2006).
Sony Ericsson bought the UIQ mobile
phone user interface from Symbian in 2006,
which has helped it develop the new Sym-
bian smartphones it showcased at MWC.
But Sony Ericsson (a 13.1 percent share-
holder in Symbian) is no longer faithful to
the Symbian cause. In a blaze of public-
ity at MWC, Sony Ericsson announced a
new premium touch-screen phone using
Microsofts mobile OS. Called the Xperia X1,
it is scheduled to launch in the second half of
this year. This is the frst time Sony Ericsson
has developed a smartphone not using the
Symbian OS.
Cliford plays down the signifcance of
Sony Ericssons move. Its not unusual for
Symbian licensees to use multiple platforms,
he says.
This is true, but isnt there an implication
that Sony Ericsson can get a more attrac-
tive and cost-efective touch-screen phone
with Microsoft (clearly aimed at competing
against iPhone) than it can with Symbian?
[X1] raises questions regarding the readi-
ness of Symbian and UIQ to power devices
with very rich user experiences,Windsor
writes, but we think that it is more about
time to market and PC connectivity.
Unsurprisingly, Cliford disagrees Symbian
is lacking from a functionality point of view.
There are no limitations on the Symbian
OS for supporting touch-screen phones,he
says. During Q105 we outsold Microsoft by
6 million units, but during 4Q07 we outsold
Microsoft by 20 million units. Its clear where
the momentum is,he continues.
Cliford says hes also not fazed by the
prospect of Googles royalty-free mobile OS,
known as Android, coming onto market.
This is the 24
th
iteration of the Linux plat-
form,he says. Its going to be very dif cult
for developers to know what to build to.
A range of heavyweight mobile opera-
tors, however, appears to be taking Android
seriously. Verizon, China Mobile, KDDI, NTT
DoCoMo, T-Mobile and Sprint are all mem-
bers of the Open Handset Alliance, which
aims to promote Android.
T-Mobile also announced at MWC it will be
launching an Android phone during 4Q08.
It will be groundbreaking,says T-Mobile
International CEO Hamid Akhavan at the
operators press conference, referring to
the anticipated improvement in the Web
browsing experience compared with other
smartphones.
If this turns out to be true, the royalty-based
model adopted by Symbian will come under
pressure sooner than Clifford would like.
Ken Wieland
There are no
limitations on the
Symbian OS for
supporting touch-
screen phones
Nigel Cliford, Symbian
Symbian unaudited 4Q07 financial highlights
2007 (4Q) 2006 (4Q) 4Q YOY
Change
2007 YTD
Symbian OS units 22.4m 14.6m 53% 20.4m
Average royalty/unit US$4.3 US$5.1

US$4.8
Royalty GP percent 96% 93%

95%
Turnover m m

m
Royalties 52.7 40.9

48.2
Consulting services 3.0 3.5

2.7
Partnering & other 0.8 1.0

1.5
56.5 45.4 24% 52.4
Source: Symbian (2008)

www.telecommagazine.com

March 2008 Telecommunications

9

Global News Analysis
MART05-NewsAnalysis.FINAL.indd 9 3/4/08 1:46:06 PM
rate pricing packages and bundles. In such a
period of rapid growth and innovation, the
GSMA believes talk of regulation is prema-
ture.
In response, Rob Barnes, head of broad-
band and mobiles at Moneysupermarket.
com says, Costs for UK customers using a
mobile abroad are among the highest in
Europe and we welcome the EC steps toward
reducing text and data roaming charges.
According to the comparison Web site, in
the UK Orange customers pay 8 per Mbps,
while customers on the 3 network pay 3 per
Mbps.
The EC position is good news for con-
sumers and it will be interesting to see how
operators react. The old adage of the Roman
LegionsSPQR (Senate for the People)
could now mean Small Proft Quick Returns.
In other words, drop the margins and go for
volume!
Kendrick Struthers-Watson
fone has done is increase the amount of data
a customer can consume from 100Mbps
to 150Mbps. Vodafones mainland Europe
customers who access the Internet while
roaming will see the charge reduced by 15
down to 60 per month. It is anticipated
UK customers will be charged below these
fgures.
The GSM Association (GSMA), represent-
ing global mobile networks, has hit back
at EC (European Commission) threats to
introduce regulation to enforce data roam-
ing price cuts. It says prices are already falling
and any move to intervene in its market by
regulators could stife innovation and stunt
development of new data oferings for
roaming.
Tom Phillips, chief government and
regulatory afairs of cer at GSMA, says in a
statement: This market is evolving rapidly
as operators develop new and innovative
services and tarif packages, including fat-
tions, learn from other industries and build
strong regulatory teams.
Thomas concludes: The good news is
both senior management and those in the
regulatory function are receptive to change
and open to the idea of learning regulatory
lessons and preferred practice from the
other sector. It is essential communications
companies take a proactive approach to
defning and agreeing on their regulatory
strategy if they are to prosper in an increas-
ingly converged world.
Kendrick Struthers-Watson
Telcos regulatory strategy at risk
SPEAKING at the Mobile World
Congress in Barcelona last month, European
telecom commissioner Viviane Reading
threw down the gauntlet to mobile
operators by saying she would impose price
cuts if they do not reduce signifcantly their
charges for roaming text messages and
mobile Internet by July 1.
Currently, the average UK price for
1Mbps of mobile data (roughly up to
200 mobile Web pages) is around 1.50.
However, UK customers are charged
almost 4.11 per Mbps for roaming data
in Europe, far higher than the European
average of 5.24 (3.90). The average cost
of sending a text message in the UK is
5.6 pence, while sending one home from
abroad costs 21 pence.
In the meantime, Vodafone preempted
the EU move by announcing Feb. 8 it would
make an efective 45 percent price reduction
for mobile data consumption. What Voda-
EC throws down data roaming gauntlet
ACCORDING TO a new report from
KPMG, telcos whose board and regulatory
functions are misaligned face risk to their
corporate regulatory strategies. KPMGs Bringing
Regulation into the Boardroom, based on
interviews with more than 60 senior telecom
executives around the globe, found senior
management and the regulatory function have
diferent priorities.
The report goes on to say the regulatory
function has traditionally been seen by
many as a home for very bright but insular
ivory tower specialists.In recent years some
successful companies have benefted hugely
from embracing the intelligence and insight
such specialists can bring to issues beyond
their own (often narrow) regulatory brief.
Senior managers are keen for the regulatory
function to react to the immediate commer-
cial challenges afecting the business and
its profts, while the regulatory professionals
are more preoccupied with the longer-term
impact of regulation on corporate strategy,
given the long-term nature of regulation.
This begs the question as to whether
these companies actually have defned regu-
latory strategy,says David Thomas, director
and head of communications regulation at
KPMG in the UK. Given the profound and
growing impact of regulation in the com-
munications sector, there is a real danger
this disconnect could lead to inadequate
regulatory policy.
The survey highlighted a considerable
diference in how the board and regulatory
teams view their relationship. Almost 60
percent of regulatory teams see themselves
as business partners to the board, but only
14 percent of senior managers share this
view, regarding regulators as specialists to be
called on when needed and not an integral
part of the business.
Worryingly, the survey indicated regula-
tory strategy is not well understood across
telecom businesses, a fnding at odds with
the views of in-house regulatory teams
that believe they have communicated
this strategy efectively. Convergence of
telecommunications and media, along with
deregulation, is blurring traditional regula-
tory boundaries and putting increased
pressure on telcos to improve communica-
10

Telecommunications March 2008



www.telecommagazine.com


Global News Analysis
Europe News Analysis
Almost 60 percent of
regulatory teams see
themselves as business
partners to the board,
but only 14 percent of
senior managers
share this view
MART05-NewsAnalysis.FINAL.indd 10 3/4/08 1:46:12 PM
p11NTTDOCOMO.indd 11 3/4/08 2:00:15 PM
Europe plays FTTH catch-up
IDATE, a France-based research and
consultancy frm, shared its latest FTTH and
FTTB statistics at the FTTH Council Europes
annual conference held in Paris last month.
At frst glance, the numbers are encourag-
ing. At the end of 2007 IDATE calculates there
were just over 1 million FTTH/B subscribers
in EU31 (EU27 plus Norway, Iceland, Swit-
zerland and Andorra), which represents 23
percent growth compared with 12 months
previously.
The number of homes and buildings
passed by fber grew much faster over the
same period, however, reaching 4.9 million (a
79 percent increase).
But compared to Asia and the United
States, Europe still lags well behind in FTTH
deployment. According to a new global
FTTH ranking, also released at the FTTH
Council Europes Paris conference (and based
on statistics gathered at the end of 2007),
Asian economies continue to outpace the
rest of the world with South Korea mov-
ing into the top slot with 31.4 percent of
households connected by FTTH, followed by
Hong Kong at 23.4 percent and Japan at 21.3
percent (see table).
A large gap separates third place Japan
from fourth place Sweden, where 7.1 percent
of homes are wired with FTTH, followed
closely by Taiwan at 6.8 percent and Norway
at 6 percent. Denmark, at 2.5 percent, oc-
cupies seventh position on the chart.
The U.S., by more than doubling its
penetration rate to 2.3 percent, moved up
three places to eighth position, followed by
two of the three countries making their frst
appearance on the chart: Slovenia at 1.8 per-
cent and Iceland at 1.5 percent. The Peoples
Republic of China moved from tenth to elev-
enth place as direct fber connections in that
country moved up slightly to 1.5 percent.
Europes lop-sided
FTTH picture
Scratching below the surface of the
European fgures released by IDATE, wor-
rying patterns persist for those wishing for
extensive and pan-European deployment.
The majority of FTTH/B subscribers (86
percent), for example, are still concentrated
in fve countries (Sweden, Italy, Norway,
Netherlands and Denmark).
Whats more, it is still municipalities and
utilities that account for most of the new
FTTH/B activity. Since mid-2005 there have
been 88 new FTTH/B initiatives, and munici-
palities and utilities accounted for 61 percent
of those.
However, IDATE reports altnets are becom-
ing much more prominent of late and now
account for around half of FTTH/B subscrib-
ers in Europe. Moreover, 60 percent of homes
and buildings in EU31 are passed with fber
courtesy of the altnets. Municipalities and
utilities account for 25 percent of homes and
buildings passed.
Since the major operators in Europe, apart
from France Telecom, Iliad, Telekom Slovenia
and Orange (in Slovakia), have yet to make
any signifcant push on FTTH/B, growth has
been patchy and generally subdued.
Besides their regulatory concerns about
hefty FTTH/B investment, some Tier 1 opera-
tors at last months FTTH Council Europe con-
ference also cited problems with marketing
as a factor holding back FTTH growth.
H
o
u
s
e
h
o
l
d

p
e
n
e
t
r
a
t
i
o
n
Source: Fiber-to-the-Home Council (February 2008)
Economies with the highest penetration of
fber to the home/Building+LAN
fber to the home
fber to the building + LAN
35%
30%
25%
20%
15%
10%
5%
0%
S
o
u
t
h

K
o
r
e
a
H
o
n
g

K
o
n
g
J
a
p
a
n
S
w
e
d
e
n
T
a
i
w
a
n
N
o
r
w
a
y
D
e
n
m
a
r
k
U
n
i
t
e
d

S
t
a
t
e
s
S
l
o
v
e
n
i
a
I
c
e
l
a
n
d
P
.
R
.

C
h
i
n
a
N
e
t
h
e
r
l
a
n
d
s
I
t
a
l
y
S
i
n
g
a
p
o
r
e
Economies with greater than 1 % household penetration
France Telecom and NeufCegetel each
report that because there is no nationwide
FTTH availability they need to have sales
teams talk to customers individually and
persuade them of the services value, which is
time consuming. France Telecom says it has
only 5 percent to 10 percent take-up in areas
where it has fber coverage.
Some operators also have marketing prob-
lems frmly of their own making. Thomas
Langer, an equity research analyst covering
the German market for WestLB, points out
that Deutsche Telekom will tell its customers
if they are eligible for its FTTN+VDSL service,
but only after they have placed an order.
Given that marketing gafe and the ream
of of-putting footnotes that usually accom-
panies Deutsche Telekoms diferent VDSL
tarif packages (telling customers about
numerous additional payments and clauses),
Langer says he is not surprised Deutsche
Telekom had only around 150,000 subscrib-
ers to its VDSL service by the end of 2007.
Ken Wieland
Altnets are becoming much more prominent
of late and now account for around half of
FTTH/B subscribers in Europe
12

Telecommunications March 2008



www.telecommagazine.com


Global News Analysis
MART05-NewsAnalysis.FINAL.indd Sec1:12 3/4/08 1:46:21 PM
How do you service thousands of
people with next-generation services?
With NECs Next-Generation Optical Network, global
carriers have the opportunity to gain high levels of technical
innovation required to support advanced business and
consumer-based applications and services.
By extending ber seamlessly from the core to the access
layer, NEC allows network operators to converge their
networks over a single optical infrastructure to provide
communities added bandwidth for enhanced video, data,
voice and mobile services.
www.necam.com
NEC Corporation 2006. NEC and the NEC logo are registered trademarks of NEC Corporation.
Empowered by Innovation is a trademark of NEC Corporation.
IT SERVICES AND SOFTWARE NETWORKING AND COMPUTING SEMICONDUCTORS IMAGING AND DISPLAYS
NEC Optical Networking
p13/NEC/M08.indd 13 3/4/08 2:06:22 PM
A glimmer of hope
for IPTV
110 percent from a year earlier. The growth
may look impressive, but its paltry compared
to 24 million digital TV users, more than 10
million of which were added during 2007.
Despite diferences in transmission and
terminals, IPTV and digital TV inevitably com-
pete for customers for content, interactive
entertainment and other value-added serv-
ices. The disparity in growth clearly indicates
where the market stands.
In addition, more than 80 percent of IPTV
users are in four locales: Shanghai, Zhejiang,
Heilongjiang and Henan. Most other places
are stuck in a regulatory muddle or fnancial
shortfall. Shanghai Media Group (SMG), a
regional cable TV operator, was the frst in
the country to work with Shanghai Telecom
on IPTV, but the same business model may
not be easily duplicated by others; some lo-
cal authorities are not even sure they want to
take the lead in IPTV given its cloudy outlook
and investment risks.
But its not all bad news. Shanghai has
stood out as an early IPTV winner. The city
boasted 220,000 IPTV users by the end of last
year, the largest single market in the country.
In addition to a pro-business environment,
the city takes credit for its relentless push for
broadband access, which reached 3 million
users (50% of all households). Broadband not
only makes IPTV accessible to more house-
holds, it gives a strong boost to Internet
services as well, such as online payment and
house surveillance.
In addition, Shanghai Telecom ofers
free IPTV setup and a free trial for certain
broadband plans. If customers continue to
use IPTV after trial, they pay a fee of only
30 yuan (US$4) a month, as long as their
broadband accounts remain open. Evidently
the low-cost strategy has worked to retain
current IPTV customers and encourage new
users to sign up. It also fts the grand plan
of Shanghai Telecom, which is trying to
transform itself from a traditional carrier to
one focusing on non-voice services, such as
the Internet and IPTV, for growth.
Other locales have found diferent ap-
proaches to kick start IPTV. In Hangzhou, the
city TV station set up a joint venture with a
telco to circumvent the ban on cross-indus-
try entry; an independent entity is also free
to work with third parties in IPTV technology,
supply and investment.
In theory, all broadband users are a po-
tential target of IPTV. SARFT once estimated
the threshold ratio would be 10% to 20%,
but the actual number is disappointing: A
meager 1.7% has made the switch so far. In
a country where cable TV covers 140 million
households with moderate charge, IPTV
advocates must come up with a good reason
to convince people to forgo cable for IPTV.
The recent government notice is certainly
good news for IPTV, but whether it will fy is
a completely diferent story as the service
must still sort out common sense issues.
For example, it is still beyond the compre-
hension of many viewers to hook up a PC
and pay for Internet bandwidth to watch
television. The quality of transmission (most
IPTV runs on ADSL) can cause the picture to
be jerky and temporarily freeze, a problem
that does not exist with cable. Finally, the
idea of paying extra to watch essentially the
same TV content has never caught on with
most consumers despite the temptation of
interactive service. IPTV may sound avant-
garde, but the road to success remains long
and bumpy.
Lin Sun is a Beijing-based consultant. He has
more than 20 years of experience with the
Chinese telecom industry. Contact him at
lsun@chinanex.com.
S
o far, IPTV has failed to take of in China.
While a large and fast-growing number
of Internet users (210 million as of year-
end 2007) would seem to indicate a fertile
land for IPTV, it has not taken root. Regula-
tory problems, as well as the dif culty of
establishing partnerships between content
providers (i.e., broadcasters) and telcos, have
slowed progress.
Incremental change may be in the air,
however, and could pull IPTV out of its
morass. In addition, telecom and broadcast-
ing industries are expected to engage more
actively, which will be a key driver for growth.
First, a bit of history to help understand
where IPTV is today. China Telecom and
China Netcom (which owned all IP backbone
and operations) initiated IPTV in China in
2005. Shortly afterward, the State Adminis-
tration of Radio, Film and Television (SARFT),
feeling the threat, declared IPTV service by
local carriers illegal on grounds of the State
Councils 1999 order No. 82, which explicitly
prohibited any form of cross entry between
telcos and broadcasters.
Until this year, No. 82 was the great wall
between two mighty industries: MII controls
the Internet and SARFT sits on a large cache
of video programming. There have been
some signs of tension softening in the past,
yet SARFT adamantly insists on being in the
drivers seat for IPTV development.
A glimmer of hope came late last Janu-
ary. The State Council issued a notice (No.
1, 2008) to encourage TV broadcasters to
utilize national telecom networks and TV
networks to provide digital TV and other
value-added telecom services. Many believe
it implies services delivered on the Internet
as well, which gives a green light to broad-
casters to develop interactive services.
This is the frst efort by the central govern-
ment to tear down the barrier it created nearly
a decade ago. However, the documents
wording is so vague it ofers no specifc
measures to spur IPTV or clarify the MII/SARFT
collaboration, and does not address a number
of unanswered questions. Can telcos begin
partnership with local TV stations? What are
the benefts and potential perils in opera-
tions, support and revenue sharing? Who will
enforce the new policy, and how?
Lack of steam
According to MII, China had 1.14 million
IPTV users by the end of 2007, an increase of
China Watch Lin Sun
14

Telecommunications March 2008



www.telecommagazine.com


China Watch Lin Sun
Shanghai has
stood out as
an early IPTV
winner. The city
boasted 220,000
IPTV users by the
end of last year,
the largest single
market in the
country
MART06-China Watch.FINAL.indd 14 3/4/08 1:44:40 PM
Fusing partnerships, sharing knowledge, exploring the latest technologies and making the
right connections are what ITU TELECOM ASIA 2008 is all about. Its the crucial ICT networking
platform for the Asian region. Join leaders of industry, governments, regulators, innovators
and visionaries to explore, discuss and shape the future of Asias ICT sector. Organized by
the International Telecommunication Union (ITU). Visit www.itu.int/asia2008
MAKE THE RIGHT CONNECTIONS
p15/ITU/M08.indd 15 3/4/08 2:07:37 PM
View from the Top Mark Kaish
16

Telecommunications March 2008



www.telecommagazine.com


Calling all subscribers
Voice is serious
business for Cox
Communications
by Jim Barthold
WHEN MARK KAISH left his
position as vice president of next-generation
services at BellSouth in April 2005, his new
company, Cox Communications, was the
countrys 11
th
largest telco. Now its number
seven with a bullet.
Were making some pretty signifcant
progress [and] I take credit for that,Kaish
says with a laugh to indicate hes not being
seriousat least about the credit part. Cox,
arguably the frst cable operator to get into
the phone business, has been growing for
some time and probably would have contin-
ued to climb the ladder with or without Kaish
aboard as vice president of voice strategy
and product management.
Of course it doesnt hurt to have a former
VoIP service leader at BellSouth to steer the
cable operators voice ship. While quick to
laugh and appreciate humor, Kaish is serious
about voice services and what they mean to
a cable industry facing true competition for
the frst time in its more than half-century
monopoly.
Weve been at this since 97 and have
coverage in 99 percent of our footprint.
The other 1 percent is rinky dink territories
where it just doesnt make sense, Kaish
remarks.
Twenty-six percent, or about 2.5 million
subscribers who fall within the footprint of
the nations fourth largest cable operator,
have fallen for voice services; and that 26
percent is absolutely growing,Kaish says.
In two of the markets we actually have
more customers in our footprint than the
[incumbent] telco does; we have over 50
percent share.
Verizons FiOS threat
Kaish has heard and seen some of his
fellow cable brethren pushing voice onto
the backburner to combat surprisingly
strong satellite and telco HD video. He also
knows Verizon is counting on FiOS TV and
broadband to staunch the steady stream of
wireline customers drifting to mobile and
cable oferings.
In its fourth quarter and year-end earn-
ings report, Verizon executives specifcally
pointed to FiOS TV and broadband as key
wireline subscriber retention tools.
The FiOS markets are stronger from that
perspective. When I look at access lines and
the trends in access lines it is absolutely
clear FiOS is having an impact. FiOS is clearly
helpful,Doreen Tobin, EVP-CFO, of Verizon
Communications told analysts during the
earnings call.
Verizon holds no exclusive on that trend,
Kaish says.
The bundle that holds
Our triple-play (i.e., voice, video and data)
customers churn at one-third the rate of our
non-bundled customers,he says.
The thing is, Cox has a triple play but
Verizon and AT&T can ofer more than just
three services since both have wireless
components. The emergence of femtocells
could make those wireless assets even more
valuable when wedging an ofer between
TV and broadband plays.
We think wireless is important [but] we
cant really talk about the plans,Kaish says
before proceeding to do just that. Were not
going to be the ffth wireless provider in a
market; were going to be the second inte-
grated provider in the market behind the
incumbent telco and in some cases were
going to be the frst.
Those cases would be where Cox butts
heads with EMBARQ or Qwest Communica-
tions, neither of which has an integrated
wireless element like Verizon or AT&T.
Then again, neither does Cox. The MSO
participated in the FCCs 700 MHz auction,
which was still going on when this article
was written, and it has been projected
as using that spectrumif it gets itfor
a form of integrated wireless ofering,
although company executives have never
P R OF I L E
MARK KAISH
EMPLOYMENT
Provides leadership for the future
development, deployment and sup-
port of Cox digital telephone initiatives
by overseeing the companys overall
voice strategy and the integration of
voice product features across all Cox
products
As vice president of next generation
solutions for BellSouth Corp, launched
MPLS-based network hosted VPN serv-
ices for businesses and designed and
executed a comprehensive enterprise
VoIP product strategy
Served as executive vice president of
hosting services and sales support for
Savvis Communications
Served four years as a supply of cer in
the U.S. Navy before entering the com-
munications industry
EDUCATION
B.S. degrees in economics and public
policy studies, Duke University; MBA
from the University of Virginia
MART07-View.FINAL.indd 16 3/4/08 1:42:57 PM
View from the Top
INTERVIEW
MARK KAISH
VP, Cox Communications
said thats where the spectrum will be used
and Kaish avoided answering 700 MHz
questions altogether pointing to an FCC
gag order on bidding companies. Without
its own spectrum, Cox is at the mercy of
Sprint Nextel, which has partnered with
cable players on a wireless play, or some
other wireless carrier.
It all comes back to video
Our feeling is we have the ability to difer-
entiate ourselves, ofer some unique service
to do some things the others cant do by
leveraging our video, phone and data assets
along with the wireless,Kaish insists.
Kaish knows wireless alone wont cut the
customer mustard; it will need to be broad-
band and it will need to be laced with attrac-
tive applications and it will need, probably, to
be linked into the fxed network.
We are very deeply into IMS,he says.
We did a technical trial last year and were
very successful. The promise of being able
to create apps from scratch, being able to
buy of-the-shelf apps, being able to match
of apps, we tested all that and it all worked.
Were very bullish on IMS as the underlying
architecture.
Cox might or might not be bullish on
femtocells as well; the normally candid
Kaish isnt saying. He did, however, point to
an interesting route the cable operator is
pursuing: DECT.
Digital enhanced cordless telecommunica-
tions is a digital wireless telephone technol-
ogy using TDMA to transmit radio signals to
cordless phones. Akin to mobile GSM, DECT
is designed especially for a smaller area with
a large number of users, such as in cities and
corporate complexes where a consumer with
a dual-mode phone could seamlessly slip
between GSM and DECT.
The new [DECT] standard will allow voice
and data streams, and the power consump-
tion, vis vis WiFi or even cell phones, theo-
retically is much better,the Cox executive
says. In Europe theyre ahead of us now, but
We are absolutely
full speed ahead,
and I dont see us
backing off at all
Mark Kaish, Cox Communications
theres some thinking about how DECT plays
into this whole wireless/wireline mobility
space.
Kaish talks like a telco pro and wears a
telco title but he also knows Coxs bread,
should it fall of the table, would land on a
side buttered by television.
Video greases the way
Video is No. 1,he says. Depending on
the customer, voice or data can be No. 2.
When were selling video were selling every-
thing else.
It works out well that Cox doesnt see
itself as a cable entertainment provider any
more than Verizon sees itself as a telephone
company or Microsoft as a software com-
pany. Everybody is doing everything to
grab and hold the now-generation
customer who will become the next-
generation consumer.
What makes Cox even more funand
enrichingfor Kaish is the companys private
status. There are no shareholders to please,
no analysts to mollify and no press to ap-
pease, fnancially speaking.
Instead of looking quarter to quarter
were looking year by year,Kaish says. Part
of the investment were making in our
infrastructure, our EON [extendable optical
network] project which is greatly increas-
ing our capacity out to the customers, were
looking multi-year.
Mini FiOS
EON, he says, is sort of a mini FiOSin that
it greatly expands Coxs fber penetration
deep into its markets.
We are absolutely full speed ahead, and
I dont see us backing of at all. We have
a tremendous opportunity here and we
need to exploit it as quickly as we can and
enhance the value of what were doing with
our customers,he concludes.
View from the Top Mark Kaish

www.telecommagazine.com

March 2008 Telecommunications

17

MART07-View.FINAL.indd 17 3/4/08 1:43:03 PM
LTE has WiMAX
in its sights
The battle for 4G intensies as LTE
supporters claim growing momentum
by Ken Wieland
The mobile & wireless opportunity LTE vs. WiMAX
18

Telecommunications March 2008



www.telecommagazine.com


LTE IS the hottest buzz-acronym in the
mobile industry. Those attending the Mobile
World Congress (MWC) extravaganza in
Barcelona last month were bombarded by
messages from a variety of vendors (and
some operators) saying the time for LTE, or
long-term evolution, is drawing ever closer.
And with momentum apparently gather-
ing for this all-IP mobile broadband technol-
ogy (dubbed 4Gin some quarters), one
possible interpretation is that time is running
short for mobile WiMAX (which purports to
occupy the 4G space now) to establish itself
as a mainstream mobile broadband technol-
ogy.
Alcatel-Lucent and NEC were among the
loudest LTE drum bangers at MWC. The two
companies announced in Barcelona they
had created a 50/50 joint venture exclusively
focused on LTE development. The companies
will combine existing R&D resources on LTE
with the aim of each releasing its frst single
LTE product line before the end of 2009.
There are two triggers for the JV with
NEC,says Alcatel-Lucent CTO Olivier Baujard.
Increased scale in R&D and market reach,
and getting commercial products quicker to
market on a global basis.
NEC is already trialing LTE with NTT
DoCoMo, while Alcatel-Lucent is trialing LTE
with Verizon. Customers have reacted very
positively [to the JV],Baujard says. They can
see it will be backed by around twice as large
R&D and system architecture capability.
The prospect of supplying NTT DoCoMo
is a key attraction for Alcatel-Lucent, which
Baujard believes will be the earliest and
biggest adopter of LTE worldwide. But NEC
benefts, too, with the chance of getting a
frmer toehold in Western markets.
The LTE product line will develop both
FDD (frequency division duplex) and TDD
(time division duplex) variants aimed at both
CDMA and WCDMA operators. Baujard is
MART08CoverStory.FINAL.indd 18 3/4/08 1:39:45 PM
The mobile & wireless opportunity
optimistic the JV will mushroom quickly. I
wouldnt be surprised if the JV with NEC, after
two or three years, had about 1,000 staf,he
says.
Ericsson also demonstrated at MWC what
it claimed to be the worlds frst end-to-end
phone call enabled by LTE. The Swedish
supplier says its LTE network equipment sup-
ports multi-user data rates of up to 160Mbps
per cell, using handheld mobile devices
developed by Ericsson Mobile Platforms.
Not so long ago, the general industry
consensus was that LTE wouldnt be around
until after 2010. But now, major suppliers
are saying the frst commercial deployments
will be much sooner. Robert Puskaric, head
of Ericssons mobile platforms business, in a
prepared statement released during MWC,
said, This world-frst demonstration marks
a key milestone in the development of LTE.
It proves the stability and integration of our
platform with a form-factor-accurate hand-
held device, reaching uplink and downlink
data speeds of 25Mbps already today. Sam-
ples of our frst LTE platform will be available
in 2008.
Jean-Pierre Bienam, chairman of the
UMTS Forum (which supports the 3GPP
family of standards, which it says includes
LTE), is equally bullish on LTE development. I
think 90 to 95 percent of LTE standardization
is already fnished,he says. I am confdent
we can fnalize the LTE standard by the mid-
dle of this year, have the frst commercial

www.telecommagazine.com

March 2008 Telecommunications

19

LTE vs. WiMAX
A big bone of contention throughout the supplier community is how far
LTE is backward compatible to 3G in comparison with mobile WiMAX. It is an
important facet of the WiMAX vs. LTE debate as it will not be possible to have
4G coverage from day one. If backwards compatibility to 3G is indeed easier
and cheaper with LTE than it is with WiMAX, that would be major argument in
favor of going for LTE (assuming an operators spectrum allocation would not
preclude either WiMAX or LTE adoption).
Jean-Pierre Bienam, chairman of the UMTS Forum, has no doubt LTE is a
more optimal path for existing 3G operators than alternative 4G technologies,
such as WiMAX. It will be possible to go directly from HSPA to LTE within two
years,he says. Its a modular solution because there is compatibility.
Those such as the UMTS Forum that propose LTE is part of the 3GPP family
of standardsargue the LTE RAN (radio access network) is being standardized as
a complement to the existing 3GPP family.
LTE parameters and procedures have been selected in such a way as
to allow co-existence, smooth migration and high performance handovers
between the existing 3GPP and 3GPP2 systems,says Alan Hadden, president of
the mobile Global Suppliers Association, which lobbies on behalf of 3GPP and
LTE. LTE terminals will from day one be dual mode and support roaming and
hand-over to currently deployed networks. LTE is backwards compatible to the
rest of the established technologies in the 3G family.
Hadden adds that standardization work is ongoing to enable handover
and interworking with other non-3GPP systems, such as CDMA2000, WiFi and
WiMAX, but in a less optimized manner.
These types of arguments are refuted strongly by the likes of Intel, the biggest
supporter of WiMAX. It argues the use of modular solutions to ensure backwards
compatibility with 3G could just as easily be implemented alongside WiMAX as LTE.
Nothing is impossible,Bienam responds, but that would take some time
and development. Backwards compatibility is easier with LTE.
According to Bienam, WiMAX will play a niche role going forward
compared to LTE (though not necessarily one that is unattractive in market
size). Weve always adopted the position that mobile WiMAX will be
complementary to 3G+ and LTE, in much the same way as WiFi is to WCDMA,
he says. WiMAX is above all a MAN network. The full coverage will be assured
by cellular technologies.
Nevertheless, Bienam hints the WiMAX opportunity will diminish once LTE
arrives. The mobile WiMAX position is for data, and 3G for voice. But operators
will not invest in LTE only for voice,he says.
deployments by 2010, and have widespread
deployments during 2011.
In January 2008, 3GPP announced some
of the RAN (radio access network) specifca-
tions had been frozen, but it was generally
believed at the time it wouldnt be until the
end of the year that the 3GPP Release 8 speci-
fcation for LTE would be fnalized.
Bienam clearly sees it diferently and
insists there has been no undue haste to
catch up with mobile WiMAX in the wake of
comments made by Arun Sarin at last years
show in Barcelona. At the 3GSM event, the
Vodafone CEO famously threw down the
gauntlet to 3G suppliers to develop data-
friendly networks as quickly as possible. He
noted that other data-centric technologies,
such as WiMAX, could eat our lunch.
Chris Beardsmore, WiMAX product market-
ing manager at Intel, which has invested
more than US$1bn in WiMAX operations
around the world, is skeptical that LTE suppli-
ers will deliver commercial networks in such
a short timeframe. The LTE standard has yet
to be fnalized and there is still a lot of work
to done,he says. We were doing the same
booth demos as LTE are doing now more
than two years ago on mobile WiMAX, and
we didnt have a standard agreed [upon] until
December 2005.
Bienam remains undaunted and adds
that the NGMN (next generation mobile
network) initiative, created last year by some
of the worlds leading mobile operators to
look at 4G options, is destined to go for LTE. I
think if you take an open view, without being
partisan, there is a major trend for converging
toward LTE,he says.
Recent announcements would seem to
support Bienams view. Telstra, AT&T and
Verizon already have committed to LTE and,
during the course of the MWC event, China
Mobile also revealed it would join Verizon
Wireless and Vodafone in coordinating LTE
Is LTE a more optimal 4G path than WiMAX?
LTE parameters and procedures have
been selected in such a way as to allow
co-existence, smooth migration and high
performance handovers between the existing
3GPP and 3GPP2 systems
Alan Hadden, GSA
MART08CoverStory.FINAL.indd 19 3/4/08 1:40:21 PM
The mobile & wireless opportunity
trials, which will focus on both FDD and
TDD modes. The TDD emphasis is signifcant
given that Chinas homegrown 3G standard,
TD-SCDMA, uses TDD. A TDD variant would
give LTE a better chance of adoption in
China as a 4G path from TD-SCDMA.
As the bulk of mobile WiMAX R&D has
gone on TDD, it hardly can be ruled out as
a 4G contender in the China market. But
the announcement by Ericsson earlier this
year that it had demonstrated LTE in both
FDD and TDD modes on the same platform
would seem to suggest the Swedish sup-
20

Telecommunications March 2008



www.telecommagazine.com


plier, which has no mobile WiMAX portfolio
of its own, is looking to compete directly
with mobile WiMAX on TDD territory.
The WiMAX Forum is believed to be
exploring FDD solutions of its own, however,
which would enable it to compete head to
head with LTE in the 4G space where opera-
tors have suitable spectrum allocations that
would allow them to go down either the LTE
or WiMAX route.
A converged 4G standard?
Vodafones Sarin, speaking at this years
MWC, called for unifed standards.His
proposal is to integrate TDD WiMAX with
FDD LTE into a single standard, which would,
theoretically, make the lives of carriers much
easier. No longer would they have to test both
WiMAX and LTE equipment separately and,
again theoretically, it would eventually reduce
infrastructure costs and optimize R&D re-
sources already allocated by LTE and WiMAX
suppliers on FDD and TDD respectively.
Technically, there appears to be no
insurmountable barrier to making a unifed
standard happen. We estimate there is
LTE vs. WiMAX
around 75 to 80 percent reuse between
WiMAX technology and LTE,says Fred Wright,
Motorola senior vice president of WiMAX and
cellular networks. The LTE equipment we
have basically uses WiMAX hardware compo-
nents, although the software is diferent.
Wright is not prepared to speculate how
long it would take to make WiMAX and LTE
interoperate in one single standard. Alcatel-
Lucents Baujard believes, however, it could
take up to several years.
If standards do converge, I think it would
be best for everyone involved,Wright
adds, but IPR issues could be a problem.
The WiMAX community has managed to
establish a distributed IPR regime where no
one company dominates. We [Motorola] are
proposing a similar IPR structure [to various
standards groups] for LTE because it is good
for the industry,Wright says. If we have
rational people out there, we can work it out.
Motorola, like Nokia Siemens Networks,
Alcatel-Lucent, Samsung and Nortel, is
developing LTE at the same time it is ship-
ping WiMAX equipment. And with a view
on the two technologies, Wright distances
himself from comments made by the
likes of the UMTS Forum that LTE is more
compatible with 3G than WiMAX (see Is
LTE a more optimal 4G path than WiMAX?
page 19).
LTE is going to be, essentially, an overlay
network,he says. It doesnt matter who the
incumbent supplier is. The vast majority of
base stations that exist today in the world of
UMTS or GSM are not upgradeable. Lets say
an Ericsson or a Nokia does have a brand new
base station that is upgradeable to LTE. What
percentage of those base stations will exist
in the global market by the time you start
rolling out LTE anyway?
At the moment, Wright says there is almost
no hardware reuse between UMTS and LTE.
However, Motorola has developed a retroft
kit to allow an LTE carrier to be installed in
Verizons existing GSM base stations. By using
the same physical package, this will help
conserve foor space and optimize power
consumption.
But, as Wright is keen to point out in all the
debate surrounding LTE and WiMAX, WiMAX
has one main advantage over its would-be
4G rival. The one thing LTE doesnt have go-
ing for it is it doesnt exist. WiMAX is here and
now,he says.
Wright says Motorola has shipped more
than 3,000 WiMAX base stations and shipped
almost 50,000 CPE units. With 16 contracts
with operators, another 60 active engage-
ments,and a WiMAX order book worth
nearly US$1.5bn, Motorolas WiMAX business
is growing impressively. WiMAX flls a very
defned need for certain operators around
the world,he says.
LTE subscribers to top 400
million by 2015, Analysys says
The worldwide LTE subscriber base will reach 437 million by 2015. This is a projection by the Analysys
frm in its recent report Global Mobile Broadband: Market Potential for 3G LTE. And while Western Europe
is expected to account for 150 million of that number, Analysys expects strong LTE growth in developing
markets. The proportion of LTE subscribers from developing regions, Analysys says, will rise from 23
percent in 2011 to 36 percent by 2015 (see table below).
LTE revenues are also expected to grow strongly, reaching around 150bn by 2015.
Of that sum, around 36 percent are expected to come from non-voice services, which is signifcantly
higher than current non-voice revenue rates of around 10 percent to 15 percent in developed markets.
500
450
400
350
300
250
200
150
100
50
0
2011 2012 2013 2014 2015
S
u
b
s
c
r
i
b
e
r
s

(
m
i
l
l
i
o
n
s
)
Source: Analysys Research (2007)
Forecast LTE subscribers by region, 2011-2015
Rest of World
Rest of Asia
Rest of the Americas
Eastern Europe
Developed Asia
North America
Western Europe
The one thing LTE
doesnt have going
for it is it doesnt
exist. WiMAX is
here and now
Fred Wright, Motorola
MART08CoverStory.FINAL.indd 20 3/4/08 1:40:32 PM
Its not surprising that
mobile providers view IP as
the key enabler for backhaul
transformation.
Ceragon. High-capacity wireless Ethernet solutions
for cost-efficient mobile backhaul.
www.ceragon.com
Weve known it
for years.
CERAGON
p21Ceragon.indd 21 3/5/08 9:23:06 AM
2007 WAS THE rst year in which mo-
bile data trafc exceeded that of voice
in many networks. This is a piece
of signicant, albeit not surprising,
news. After all, we no longer think of
our cellular handset as simply a mo-
bile phone but rather as an extension
of our PC. Sometimes even as its re-
placement. Web browsing, e-mail ex-
change, music downloading and video
sharing while on the go, are only a few
examples of the wide variety of servic-
es offered to us via our mobile devices.
And judging by the current onslaught
of new high-end handsets, this trend
will only intensify in the near future.
In a perfect world, operators would
be rejoicing over the growing volumes
of trafc streaming back and forth
through their mobile networks. But
in reality, this growth poses one of the
biggest challenges mobile service pro-
viders have ever faced. This is because
the increase in data-centric communi-
cation is coupled with revenue erosion
per transported bit. In other words,
not only are operators not making ad-
ditional revenue from the excess use of
their networks-they actually have to
struggle to maintain protability.
Nowhere in the network is the ef-
fect of increased data trafc more sig-
nicant than in the backhaul segment.
Connecting between cell towers and
the operators core network, backhaul
is a vital link in the mobile operators
communication chain. It is also where
operators are currently focusing their
efforts to decouple cost and capacity
by taking these important measures.
First, increase capacity across the en-
tire backhaul network-from tail sites,
through chain to aggregation sites-to
cope with the additional demand for
bandwidth. Second, speed up the mi-
gration to all-IP networks to benet
from the exibility and higher net-
work efciency of IP and Ethernet so-
lutions-which also offer signicantly
lower cost-per-bit compared with
legacy TDM. Finally, a shift toward
advanced wireless backhaul solutions
that support much higher data rates
than copper lines, while overcoming
Broadband in the air
How new wireless backhaul solutions help to reduce the cost and
complexity of next-generation mobile networks
the cost and limited avail-
ability of ber. Figure 1
depicts a typical backhaul
network architecture.
Capacity, capacity,
capacity
It has already become
a worn-out clich, but
the demand for more
bandwidth just keeps on
growing. For example,
3/3.5G phones make up
only about 10 percent of
the handsets in use today.
Yet, in some parts of the world, mobile
networks are already strained to the
limit supporting data-rich applications
and services. In Europe, Deutsche
Telekoms CEO Ren Obermann was
quoted as saying that on average, DTs
iPhone subscribers use 30 times more
bandwidth than average contract-
based customers. Imagine how much
more bandwidth theyll consume when
iPhone goes 3G.
And its not only mobile phones that
are generating data trafc-its PCs too.
3G access dongles are becoming ever
more popular, making 3/3.5G net-
works a true alternative to traditional
xed-line broadband networks. 3G
broadband access is comparable with
most DSL services, with one major ad-
vantage: mobility. All this boils down
to one thing: capacity.
The hunger for bandwidth does not
go unnoticed. Mobile operators world-
wide are already considering ways to
enhance their backhaul capacity. A
single tail-site today is
typically fed by 4Mbps
to 8Mbps. As shown in
Figure 2, this is expected
to quadruple over the
next two to ve years and
trigger a chain reaction
of upgrades across the
entire network. WiMAX
base stations, planned
with 20Mbps to 30Mbps
backhaul from the get-go,
will require even higher
capacities farther down
the network. As currently installed
tail-site backhaul solutions, both wired
and wireless, start to max out, opera-
tors need to look for solutions that can
provide a wider range of capacities.
Such solutions are the only way to en-
sure continued subscriber and service
expansion without having to reinvest
in equipment upgrades every two to
three years.
Enabling smooth migration to
all-IP backhaul networks
So how can operators achieve their
capacity goals quickly and cost effec-
tively? Selecting exible solutions with
enhanced bandwidth versatility is the
rst step. Migrating their networks to
all-IP is the second.
Networks based on IP or IP/Ether-
net are less expensive to operate and
maintain than legacy TDM. IP also
allows for better exibility and truly
efcient network utilization, helping
to decouple cost and capacity. In ad-
By Aviv Ronai, Chief Marketing Ofcer, Ceragon Networks
TechFocus on
MOBILE BACKHAUL
Figure 1. Microwave backhaul network topology example
Figure 2. Capacity at the cell-site is on the rise
p22-23CERAGON ad.indd 22 3/4/08 2:21:27 PM
dition, IP/Ethernet is taking over as
the transport technology of choice
because it allows efcient streaming of
advanced applications, such as mobile
TV, which will be the future revenue
generators for cellular operators. IP
networks are now also geared to deliver
carrier-grade services using Ethernet
transport. MPLS (multiprotocol la-
bel switching) and PBB-TE (provider
backbone bridge-trafc engineering)
allow operators to manage data paths
within large networks supporting hard
QoS, 50ms resiliency and carrier-class
OAM tools in Ethernet environments.
The IP migration phase will take
several years. During this time hy-
brid networks, in which TDM and
IP equipment coexist, will dominate
the market. One way of ensuring a
smooth migration and successful mi-
gration architectures is by employing
the Native
2
concept. Also referred to
as ofoad or hybrid, Native
2
sug-
gests a gradual IP/Ethernet phase-in
with a later phase-out of TDM. By
delivering native Ethernet alongside
native TDM, operators can maintain
two different transport paths: TDM
for voice; IP/Ethernet for data. This
does not suggest using two networks,
but rather having a single platform
that supports both native TDM and
native Ethernet. In other words, Na-
tive
2
means no future replacement of
equipment is required, capping legacy
investments and protecting capex for
years to come.
Reducing deployment time,
complexity & costs
Point-to-point microwave backhaul
solutions are ideal for all segments of
the operators backhaul network. Un-
like copper lines (T1/E1) that have a
limited, soon-to-be-maxed-out capac-
ity, microwave links have a broad ca-
pacity range, from as low as 10Mbps
to as high as 1Gbps over a single radio
channel. Unlike ber, which
has virtually limitless capac-
ity but is highly expensive and
limited in availability, micro-
wave solutions can be set up
and put into operation in a
matter of hours.
New ACM techniques now
help maximize spectral utili-
zation in IP/Ethernet-based
microwave systems. This tech-
nological enhancement allows
microwave systems to provide
higher bandwidths, over great-
er distances, using smaller an-
tennas. ACM is in essence an
automatic adjustments that a
wireless system can make to prevent
weather-related fading from causing
communication disruptions on the
link. When heavy weather conditions,
such as a storm, affect the transmission
and reception of data and voice over
the wireless network, the radio system
automatically changes the modulation.
By doing so, real-time applications
(such as voice and video) will always
continue to run smoothly.
To better understand how ACM
works to benet mobile operators, lets
look at the following example depicted
in Figure 3, which compares a legacy
high-capacity TDM solution with a
native Ethernet solution deploying
ACM. In this particular case, a DS3
service delivered over a 50Mbps link
at 99.999 percent availability. Alter-
natively, the same system can provide
a single service carrying 155Mbps at
99.99 percent availability, delivering
STM-1/OC-3 service.
On the other hand, a native Ethernet
link using the same frequency spectrum
and enhanced with adaptive modulation
offers not one but a series of services in-
cluding: carrier-grade 50Mbps at 99.999
percent availability; a second service pro-
viding 105Mbps at 99.99 percent avail-
ability; and a third service delivering an
additional 45Mbps at 99.9
percent availability, for a
total of 200Mbps. In other
words, a link planned for
50Mbps at 99.999 per-
cent availability yields ad-
ditional 150Mbps free
capacity for non-real-time
data services with no extra
investment on spectrum
or hardware, and using
the same exact link budget
and antennas.
ACM also helps op-
erators reduce costs-as shown by the
example depicted in Figure 4. In this
case, we can see that a microwave sys-
tem powered with ACM can make use
of 1-foot antennas with minimal trade-
off, as compared with a system using
4-foot antennas without. In this case,
an operator can guarantee a 100Mbps
link at 99.999 percent availablity for
real-time, delay sensitive services,
while offering sufcient four-nines
and three-nines availablity for an ad-
ditional 300Mbps of non-real-time ap-
plications.
As shown in the example, outage
time is identical for both systems. But,
using smaller antennas can amount to
big savings. Roof and tower space lease
in the United States is on average $100
per foot per month. A simple calcu-
lation shows that an operator would
have to pay $4,800 a year for a 4-foot
antenna compared with only $1,200
for a 1-foot. Thats $3,600 saved every
year per antenna, or $7,200 per link.
Conclusion
The demand for bandwidth shows
no signs of slowing down. Further
down the road, 4G technologies such
as WiMAX and LTE, expected to be-
come commercial in two or three years,
will bring the use of data applications
over mobile networks to new peaks.
As mobile networks become more
and more data-centric, mobile opera-
tors need to seek ways to increase ca-
pacity without a linear increase in cost.
On the backhaul side, high-capacity
microwave solutions that support na-
tive IP/Ethernet trafc and feature
advanced adaptive modulation tech-
niques for maximum spectrum utiliza-
tion are helping them do just that.
Aviv Ronai is chief marketing ofcer at
Ceragon Networks.
Figure 3. Ethernet microwave with adaptive modulation: gain overall
capacity while increasing availability
Figure 4. Ethernet microwave with adaptive modulation:
similar performance, lower costs
p22-23CERAGON ad.indd 23 3/4/08 2:21:32 PM
3G awaits
Olympic start
The mobile & wireless opportunity China
merit or expectation of new service, but as
a symbol of Chinas rising power in world
telecommunications. From the outset, TD-
SCDMA was never merely a market issue but
instead charged with national pride, political
correctness and frm government support.
To a great extent, this has propelled rapid
development, mainly in handset design and
performance issues such as video display
in high-speed moving vehicles, antenna,
battery life and compatibility with GSM
(dual-mode operations for roaming). More
important to its future growth, TD-SCDMA
has spawned in two short years a new
ecosystem, moving from operating system
to video module, chipset to base station, and
lab test to feld trial.
What to expect
The Chinese government has played a vital
role in bringing TD-SCDMA to fruition from a
mere crude spec. In fact, the government has
delayed 3G licensing in part to allow more
time for TD-SCDMA to catch up with compet-
ing standards. The Olympic Games ofer the
perfect occasion to debut TD-SCDMA while
excluding WCDMA and CDMA.
Many criticize the Chinese government for
its anti-market practice, but this is how the
government enforces its determination to
reverse an otherwise normal 3G evolution.
The Chinese governments pledge to
launch 3G in Beijing this summer puts
TD-SCDMA, the countrys homegrown
standard, under the spotlight. Can it
deliver?
Lin Sun
24

Telecommunications March 2008



www.telecommagazine.com


WHEN THE Ministry of Information
Industry (MII) pledged in 2006 that China
would launch 3G service during this sum-
mers Olympic Games in Beijing, the promise
brought mixed feelings. On one hand, it
rekindled hope the Chinese 3G standard, TD-
SCDMA, would fnally take center stage. On
the other, the pledge was a serious challenge
to an unproven technology facing competition
from veteran standards WCDMA and CDMA
EV-DO in the worlds largest cell phone market.
Whatever the outcome, 3G service will open
a new chapter in mobile communications in
China; one with a very diferent landscape.
As the Olympics draw near, many look
eagerly to TD-SCDMA, not on its technical
MART12-Feature.FINAL.indd 24 3/4/08 1:35:36 PM
China Mobile is the largest GSM operator in
the world; it would adopt WCDMA as a logi-
cal migration to 3G under normal circum-
stances. However, the operator was directed
to develop most sites for the Olympics with
TD-SCDMA: eight cities out of 10.
Although this may not necessarily be a
bitter pill for China Mobile to swallowit can
preempt all competition as a de facto TD-
SCDMA operator if it turns out to be success-
ful, plus it still can ofer dual-mode service for
its GSM customersthe operator must forgo
a preferred path and devote its resources to a
new standard.
Apparently the government does not
think that way. In its eye, China Mobile is the
best candidate for TD-SCDMA: It runs the
worlds largest cell phone network, has expe-
rience with large-scale operations, and sits on
a large hoard of cash. It is the best assurance
China has to make TD-SCDMA a success.
Despite the fanfare, what will happen
during and after the Olympics is anybodys
guess. The Games will take place in 10 cities
and only last for two weeks during which
most use of 3G service will be limited to
people associated with the sporting event:
of cials, athletes, coaches, reporters and
those dealing with logistics. If travel occurs
outside the venue, handsets will switch au-
tomatically to GSM which does not support
high-speed data service. Clearly the Olympics
is not a good gauge of 3G demand in China,
not only because service will be limited, but
also because handsets will be ofered as free
rental and most services will be free of charge
as a show of good will.
Post-Olympic excitement
But the real excitement is expected after
the Olympics. China Mobile plans to install
15,000 BTSs (a 10-million-user capac-
ity)way beyond the sporting events
needs. Meanwhile, TD-SCDMA networks are
under construction in a dozen more cities in
preparation for quick launch once TD-SCDMA
receives the green light for commercial
service. Initial growth will depend on service
availability, cost of handsets, basic service,
and teaser ofers, incentives and bundling
that are still on the drawing board.
It is estimated the number of cell phone
users will continue to grow an average of 54
million a year and reach nearly 900 million by
2013. During that time 3G users will rise from
0.5 percent in 2008 to 25 percent. The acceler-
ated growth is attributed to market momen-
tum, falling prices and competition from other
3G services.
Another critical factor afecting TD-SCDMAs
future and the overall 3G market is operator
restructuring. China Unicom seems likely to
merge with China Netcom sometime this
yuan (US$3.5bn). So far, most shipments
have been delivered and at least six cities
have completed installation. In Beijing, the
largest Olympic site, the TD-SCDMA net-
work not only covers all stadiums and the
Olympic Village, but also is being extended
to the citys fnancial district and govern-
ment quarters.
On the outside, however, you dont see
much going on because the project still
is considered a trial without a commercial
license; operators want to keep a low profle
because no one is certain it will be a success.
Obviously China does not want to be embar-
rassed in front of the world by touting TD-
SCDMA if it turns out to be a fop. Regarding
licensing, the government apparently wants
to wait until after the Olympics to decide
when and how to proceed.
The mobilecom market
Licensing aside, the real efect of 3G on the
mobilecom market may have been overblown.
Many are carried away by 3Gs charm and have
made egregious predictions of what 3G may
become in China. Most predictions assume
3G will replace current mobile service in grand
scale, which is unlikely to happen. At best, 3G
will supplement where 2G falls short by ofering
an option for customers who demand it: mainly
video content (e.g., video phone, live TV broad-
cast, video games).
It is intriguing to watch TV on a slick cell
phone screen, but not everyone has the
desire or willingness to pay for the service.
The majority of people most likely will stick
with what they do today, namely voice, text
messaging, ring tones, music downloads and
Internet access, even after 3G. To them, 3G is
not a must-have but a fancy option. Making
the switch will be determined largely by
handset price and service charge.
3G will take of gradually in China forming
a critical mass rather than demonstrating
explosive growth. This will afect operators
in two ways. First, spending on network buil-
dout will be incremental and revenue from
3G will be less rosy. Second, operators will
continue to rely on current service for growth
by recruiting customers and improving ARPU
(see 3G revenue outlook, page 24).
Money in handsets
If the past is any guide, a signifcant amount
of 3G revenue will come from handset sales.
By one estimate, 3G handsets will account for
year and sell its CDMA operations to China
Telecom. The change, if confrmed, will
reduce the number of 3G licenses to three:
China Mobile for TD-SCDMA; China Unicom
for WCDMA; China Telecom for CDMA EV-DO.
Such division clearly refects the govern-
ments intention to seek balance among 3G
technologies and operator strengths. The
largest and most experienced operator, China
Mobile, will undertake the least proven stand-
ard, while two smaller competitors will carry
more mature services.
The government, however, does not want
to let on whos in line to receive a license
and with which standard. It is premature to
assume all operators currently deploying
TD-SCDMA networks will receive permission
to operate one. In fact, licensing is more tied
to restructuring than to 3G trials. Some ana-
lysts favor TD-SCDMA given its momentum
and government support and estimate the
Chinese standard will hold the lions share (50
percent) by 2010, followed by WCDMA with
40 percent and CDMA EV-DO with10 percent,
but market dynamicsafected by competi-
tioncan change.
Government control
Another cloud hanging over 3G is license
type. At this time, it is unclear if the license
will be general and allow licensees to decide
which standard to use or whether the
government will require a specifc standard.
The Chinese government has promised the
WTO and foreign governments it would not
interfere with market activityin choosing a
3G standard.
Most people, however, think the govern-
ment tells the world what it wants to hear.
The Chinese government can exercise its in-
fuence in many ways: blatantly, in the form
of notice and executive order; or implicitly
with tax, fnance and subsidy. There is little
clue as to what the government intends to
do on this issue. Since late 2007, the guide-
line for reform from government documents
and of cials has been to accelerate full
service operators. Full service, however,
usually refers to eliminating restrictions
between fxed and mobile services, not to
giving an operator freedom to choose what
it deems ft.
Progress is well underway on the ground.
Last March, three operators (China Mobile,
China Telecom and China Netcom) ordered
TD-SCDMA equipment worth 27 billion

www.telecommagazine.com

March 2008 Telecommunications

25

The mobile & wireless opportunity China
Obviously China does not want to be
embarrassed in front of the world by touting
TD-SCDMA if it turns out to be a op
MART12-Feature.FINAL.indd 25 3/4/08 1:35:50 PM
26

Telecommunications March 2008



www.telecommagazine.com


The mobile & wireless opportunity
ITUs recent ratifcation of WiMAX as an extension to IMT-2000 has
generated little excitement in China. A great beneft to mobile data service,
WiMAX supplements the traditional 3G platform which has limited capacity
for data (typically 384kbps or lower). In Chinas case, however, the market
must frst identify strong demand for high-speed data apps to justify hefty
investment in a public service. In fact, China Mobile and China Unicom have
conducted small-scale WiMAX trials in cities such as Beijing, Shanghai and
Guangzhou with service limited to local businesses, residents and schools,
but they do not plan to expand trials or envision WiMAX as part of their 3G
oferings.
In the meantime, six cities are beginning to build out WiMAX, not as a
public service but as part of a city management program called Wireless City.
Shanghai, for example, is building a mesh WiFi/WiMAX network in one county
that uses video cameras to monitor traf c and dispatch ambulances. The service
ofers 512kbps to 3Mbps bandwidth for video transmission from fxed locations
and distances. Whats interesting is these programs are built by third-party
companies in partnership with the cities, not by public operators like China
Mobile; they are not for proft and have very limited scope of operations.
Despite its superior performance over traditional 3G, WiMAX development
poses two dilemmas for China. First, WiMAX requires a standalone infrastructure
for large area coverage and high-speed data throughput, which means
operators must build two separate 3G networks: one for voice and low-
speed data; the other for those wanting special data service. This becomes a
distraction when operators are trying to sort out 3G standards and networks are
not fully built.
Voice still tops data
Again, voice will remain the most popular 3G service in China. Demand for
high-speed data beyond 384kbps will be very limited, most likely appealing
to a small number of business users, which hardly justifes WiMAX as a public
network. Even if demand for high-speed data apps rises after 3G, operators
still have the option to add HSDPA which is designed to ofer 2.8Mbps for TD-
SCDMA, suf cient for most services.
Diferent from WiMAX, HSDPA is essentially a software-enabled service
based on WCDMA infrastructure which is also the core of TD-SCDMA. Although
HSDPA does not transmit signals over long distance, it is easy to deploy, costs
much less, and shares many resources of a 3G network such as management
and provisioning. It is a more cost-efective option for 3G operators.
Second, China has yet to allocate radio spectrum for WiMAX service.
While many countries favor 2.5GHZ to 2.6GHz for WiMAX, the band is simply
not available in China. According to MIIs Radio Administration, 2300MHz to
2400MHz has been assigned as an extension to 3G TDD service; 2400 MHz to
2483.5MHz for public services like WiFi, wireless access, bluetooth devices, ISM
and other point-to-point communications; 2500 MHz to 2690MHz designated
by ITU as extended 3G spectrum. For the 3.5GHz band, 3300 MHz to 3400MHz
is reserved for GPS and other fxed and mobile access; there is about 137.5MHz
of free space in 3400MHz to 3600MHz, but it is intended for extension in fxed
wireless access.
Most current WiMAX trials operate in 3.5GHz even though the government
has never assigned the band to the service. Operators prefer to use lower
frequency for WiMAX because at a higher band data throughput is lower and
vulnerable to interference, which will afect traf c and coverage, especially
performance in high-speed motion. In addition, a WiMAX signal has less
strength to penetrate buildings and walls at higher frequency, so operators must
increase density of base stations to improve coverage, which will inevitably raise
network cost.
Another issue in spectrum is capacity. The standard 802.16e, for example,
needs at least 20MHz bandwidth per carrier to achieve a data rate of 15Mbps
to 60Mbps at speed of 150kms/h. An operator may need at least seven carriers
to cover a large area, and total bandwidth multiplies if there are three operators
ofering similar service. There have been discussions on what the government
can do to reconcile conficts in spectrum. Whatever the outcome, it will require
major changes in current policy, inevitably afecting existing and future services.
This is why the government is reluctant to act quickly in favor of WiMAX before
achieving a clear understanding of the service and its impact on other 3G
services.
Licensing hurdles
Spectrum scarcity is not the only deadlock for WiMAX; licensing is another
hurdle. WiMAX has two favors (i.e., 802.16d and 802.16e) intended for
diferent types of service and customers. Since customers using 802.16d can
roam within a coverage area, it is dif cult to diferentiate between two services
for licensing purposes and determine which operator should receive which type
of license.
Another issue is whether to license WiMAX as a public service, private
service or both. If it is a supplementary service to 3G, it could use the reserved
spectrum by licensing public operators like China Mobile. If WiMAX is used in
a closed environment like city government, enterprise or campus, it could use
high spectrum for limited coverage under a special license given to non-
public operators. A fnal issue is how to prevent unfair use of spectrum and
competition.
These issues will likely hold back WiMAX deployment in China, a market
with a large user base dominated by voice and low-speed apps. To a large
extent, WiMAX is another case of great technology looking for demand. Only
time will tell if WiMAX will fnd its niche and attain growth momentum in the
worlds largest mobilecom market.
WiMAX: a great technology looking for demand
60 percent to 70 percent of total equipment
sales. If during the next fve years 25 percent
of current cell phone users buy a 3G handset
for 2,500 yuan, that is a staggering 340 billion
yuan (US$47 billion) for handset manufac-
turers and operators that does not include
charges for service and bandwidth.
This is why operators are wary about
TD-SCDMA handsets and want to get it right
from price to performance and from design
to supply. It is also the single largest motiva-
tion for handset manufacturers to chase
operators with their latest gear: a chance to
be on the short list. China Mobile plans to
spend 4 billion to 6 billion yuan (US$526m to
China
MART12-Feature.FINAL.indd 26 3/4/08 1:35:57 PM
The IEC announces
Olympia National Hall
LONDON, ENGLAND
CONFERENCE 28 April 1 May 2008 | EXHIBITION 29 April 30 April 2008
International Engineering
Consortium
www.iec.org
International Engineering
Consortium
www.iec.org
CONFERENCE CHAIRMAN
MATT BROSS
Group Chief Technology Officer
BT Group
SOFTWARE + NETWORK =
= INNOVATION
For more information, visit iec.org/events
OFFICIAL HOST SPONSOR
SOFNET is the premier conference and
exhibition uniting software, next-generation
technologies, and service providers.
Oracle BT France Telecom Group Microsoft
Telekom Austria Hewlett-Packard Telecom Italia
Allot IBM Nexaweb Comverse Telcordia
Mindspeed Technologies Sun Microsystems Fujitsu
Keymile Huawei Italtel Meriton Schmid
p27Sofnet.indd 27 3/4/08 2:32:57 PM
US$790m) for 2 million TD-SCDMA handsets
to keep unit price at about 2,000 yuan to
3,000 yuan (US$260 to US$395), a range most
customers are willing to spend.
Because handsets are the most important
gauge for 3G growth, operators insist all
TD-SCDMA handsets be compatible with
GSM, so 3G wont become an island with little
connection to most other users. During a
recent MII test, only seven models (from sup-
pliers including ZTE, Samsung, LG, Hisense
and Lenovo) were chosen from more than
100 submissions to become frst suppliers for
China Mobile.
Tests show their products performed bet-
ter in video display, auto-switch, auto-roam-
ing, high-speed Internet access and battery
life. A Hisense model, for example, lasted
450 minutes in continuous talk time and 270
hours in standby. As 3G service becomes
readily available, handset sales are expected
to follow one of two routes: at a discount
price bundled with service plans, or through
retail channels where customers choose their
own models and services.
Video bright spot
While it is not clear how many people will
sign up for 3G, few doubt video will be a
bright spot for the new mobile service, espe-
cially with young consumers. ABI Research
predicts mobile TV users in Asia-Pacifc will
soar to 260 million by 2012, making it the
largest market in the world.
In China, some analysts estimate 20 per-
cent of 3G users or 30 million people could
use mobile TV by 2012: that translates to 10
billion yuan (US$1.3 billion) a year in service
revenue. Public sentiment seems to support
the outlook. Surveys show more than 70
percent of cell phone users would like to try
3G just for its video features if the service is
afordable. The concern, however, is 3G is not
the only way to deliver video entertainment
to handsets; in fact, it may not be the best
way compared to a more humble but efec-
tive technology.
The telecom industry has been embroiled
in drawn-out debate on a mobile TV standard
with proposals for DVB-H (Europe), Media-
FLO (United States), T-DMB (Korea) and
Chinese favors like DMB-TH (a digital TV spec
modifed for handhelds developed by Tsing-
hua University), CDMB (recommended by the
National Standardization Committee), T-MMB
(Nufrontsoft) and CMB (Huawei).
Despite certain diferences, all the propos-
als follow the same path of in-net transmis-
sion with a video overlay on top of 3G infra-
structure. China Mobile, meanwhile, requires
all TD-SCDMA handsets to be equipped with
MBMS (multimedia broadcasting multicast
system) developed by Datang Mobile for
video download and display.
is compatible with DAB, as the standard for
mobile TV. SARFT claims CMMB is completely
homegrown, saving at least 1 billion yuan
(US$130 million) in foreign IPR dues. SARFT
hopes the savings will translate to lower price
for handsets and service.
As part of its ambitious plan, SARFT has
also unveiled STiMi (satellite and terrestrial
interactive multimedia), a transmission sys-
tem comprised of satellite and terrestrial
boosters. STiMi supports bandwidth up to
7.5MHz and ofers a data rate from 2.7Mbps
Broadcasters tread quietly
While the telecom industry wrangles
over a mobile TV standard, the broadcasting
industry has launched its mobile TV service
quietly. In late 2006, Beijing Radio became
the frst in the country to broadcast two TV
channels and 12 radio stations to DAB-ena-
bled handsets and other devices via over-
the-air reception with a video module and
enhanced processing power.
A year earlier, SARFT (State Administration
of Radio, Film and TV) introduced CMMB (Chi-
na Mobile Multimedia Broadcasting), which
250
200
150
100
50
0
2008 2009 2010 2011 2012 2013
R
M
B

Y
u
a
n

(
b
i
l
l
i
o
n
s
)
Including revenue from all 3G services but excluding handset sales. $1=7.2 Yuan (as of January 2008)
3G revenue outlook
73.50
99.96
134.11
176.33
227.32
1000
900
800
700
600
500
400
300
200
100
0
T
o
t
a
l

c
e
l
l
p
h
o
n
e

u
s
e
r
s

(
m
i
l
l
i
o
n
s
)
2G/2.5G users include GSM, GPRS, CDMA and CDMA 1x.
3G user growth
Year
3.5
49.0
83.3
116.6
146.9
174.9
604.0 613.2 631.2 648.6 668.0 697.1
2008 2009 2010 2011 2012 2013
2G/2.5G users
3G users
(China, continued on page 32)
28

Telecommunications March 2008



www.telecommagazine.com


The mobile & wireless opportunity China
MART12-Feature.FINAL.indd 28 3/4/08 1:36:06 PM
MAKE THE RIGHT CONNECTIONS
Creating ideas, spreading knowledge and making the right connections are what
ITU TELECOM AFRICA 2008 is all about. Its the crucial ICT networking platform for the
African region. Join leaders of industry, governments, regulators, innovators and
visionaries to explore, discuss and shape the future of Africas ICT sector. Organized by
the I nternati onal Tel ecommuni cati on Uni on (I TU). Vi si t www.i tu.i nt/afri ca2008
p29/ITU MakeTheright.indd 29 3/4/08 2:36:04 PM
The mobile & wireless opportunity FMC
30

Telecommunications March 2008



www.telecommagazine.com


The new
mobile home
Fixed-mobile convergence has been a
disappointment until now. Could
femtocells be the answer?
by Iain Morris
AN INDUSTRY that sometimes
seems awash with unproven technologies
recently has acquired another. Almost
unheard of 18 months ago, femtocells
quickly are becoming the favor of early
2008, outscoring even WiMAX on sheer hype.
On a cursory appraisal, it may be hard to
see why. Tiny indoor base stations, femtocells
originally were designed to boost indoor cel-
lular coverage in areas where mobile phone
users still struggle to fnd a signal. Compared
to WiMAXa technology proponents hope
will revolutionize the mobile data busi-
nessfemtocells would appear to have
limited appeal.
But that perhaps misses the real story.
Femtocells also could allow an operator to
route a mobile connection indoors over a
fxed broadband access network (usually
DSL), instead of sending it to an outdoor cel-
lular base station, as typically would happen.
That might sound uninspiring, but carrying
traf c on DSL is said to be much cheaper
than using cellular infrastructure. An operator
might decide to invest in femtocells with
those cost savings in mind. It could even pass
savings on to its customers to attract new
business and hasten the trend of fxed-mo-
bile substitution.
Whats more, because femtocells can link
mobile and fxed infrastructures in this man-
ner, they also could herald the introduction
of new converged services that can jump
seamlessly from one network to another
depending on a users circumstances.
So far, other technologies that fall under the
umbrella of so-called fxed-mobile conver-
gence (FMC) have failed to live up to expecta-
tions. Could femtocells buck the trend?
The invisible femtocell
For now, mobile operators seem to be
drawn to femtocells because they want to
save money. Indeed, a compelling cost case
for investment in femtocells has been made
recently by Andrew Parkin-White, a principal
analyst with Analysys Research, who argues
femtocells represent a much cheaper alterna-
tive to outdoor infrastructure for providing
indoor coverage.
Parkin-White says his modeling indicates
the biggest benefciaries of femtocells
could be 3G operators with a relatively small
number of customers. By deploying femto-
cells in 60 percent of its customers house-
holds, such an operator potentially could
save up to US$45 per customer per year by
2012, he says. But cost savings could tail of
dramatically if deployment is less widespread.
Serving just 20 percent of households would
save only US$20 per customer per year,
according to Parkin-Whites calculations (see
Annual cost savings per customer, page 31).
MART09-Feature.FINAL.indd 30 3/4/08 1:30:05 PM

www.telecommagazine.com

March 2008 Telecommunications

31

The danger is mobile operators that fail to
adopt a large-scale approach fnd themselves
expending great efort on integrating a large
number of 3G femtocells without avoiding
signifcant investment in outdoor networks,
Parkin-White says.
Of course, cost savings were also a prime
motivation behind the launch of another
FMC technology that has not succeeded.
In the UK, BT Fusiona dual-mode phone
that switches between DSL and mobile
networkshas just 50,000 users more than
two years after launch. At press time, rumors
swirled that BT had fnally given up on
marketing the service. Only 10,000 customers
were using Deutsche Telekoms equivalent
T-One service by May 2007, almost a year
after it was introduced, at which point it was
withdrawn. Even France Telecoms Unik, often
cited as an FMC success story, had a relatively
small user base of 298,000 by June 2007,
some nine months after launch, representing
just 1.3 percent of France Telecoms mobile
customers and 1.1 percent of its fxed-line
customers. Why should femtocells do any
better?
One reason might be all those aforemen-
tioned FMC services have required handsets
to be compatible with a technology called
unlicensed mobile access (UMA), used for
transferring connections between mobile
and fxed networks. And only a narrow
range of UMA-compatible handsets is cur-
rently available. Steve Shaw, associate vice
president for Kineto Wireless which develops
UMA technology, thinks only 18 are now on
the market, although he expects this number
to rise to 40 by the end of the year. Nearly
1 billion handsets will be sold in 2008, and
we admittedly have some way to go before
every one ships with UMA,he says.
Some femtocells also use UMA to hand
mobile calls of to the DSL network, but the
technology is embedded in the base station
itself. Others rely on alternative standards.
Either way, customers can take advantage
of femtocells using their existing cellular
handsets.
Nevertheless, big challenges await
operators that see femtocells exclusively as a
means of reducing their costsnot least the
price of the femtocell itself. Arun Sarin, CEO of
the UKs Vodafone, has reportedly said femto-
cells will not be commercially viable until the
cost per unit drops to around US$100. But the
femtocells deployed in the United States by
Sprint Nextelthe only operator in the world
to have used them commerciallyprobably
cost around US$300 each, according to Dean
Bubley, an analyst with Disruptive Wireless.
Parkin-White thinks manufacturers will fnd it
immensely challengingto meet the US$100
price in the short term. At least one femtocell
ing customers femtocells are worth the
expense. In the U.S., where indoor cellular
coverage is notoriously bad, Sprint Nextel is
charging its customers US$50 for a femtocell
(leaving it to bear US$250 of the cost, unless,
of course, it has struck some kind of deal with
Samsung, the manufacturer). But in other
markets, persuading subscribers they are
worth even that could prove dif cult.
Opportunities are going to be confned to
particular groups of customers,Parkin-White
says. Nordic countries, for example, have ex-
cellent in-building coverage, and thats going
to severely limit the potential. Its no easy task
to segment the market on this basis.
Femtocells also might lose their appeal
as a coverage booster if operators continue
to roll out networks using lower-frequency
spectrum that better penetrates buildings. In
the U.S., the FCC, which regulates the com-
munications industry, is currently auctioning
700MHz spectrum formerly used by the
broadcasting industry. At press time, bidders
(thought to include mobile operators AT&T
and Verizon as well as online giant Google)
had ofered US$18.9bn, although the FCC
previously had estimated the auction would
raise US$15bn at most. The demand indicates
how much value is attached to this spectrum,
which would allow an operator to provide a
high-quality indoor service at a fraction of its
current costs.
The mobile & wireless opportunity FMC
manufacturer sounds cautious when ofering
opinion on Sarins target.
Essentially, US$100 is a good head-
line-grabbing stake in the ground,says
Mark Keenan, EMEA general manager for
RadioFrame Networks, which is developing
femtocell products in collaboration with
Nokia Siemens Networks. I dont think its too
far from what could be achieved in the future
with ef cient manufacturing and especially
high volumes. Having said that, I think the
business works with signifcantly higher
femtocell prices: They allow an operator to
save on network build and generate more
revenue from customers.
But even if costs do fall to a comfortable
level for most operators, there is the problem
of getting femtocells unobtrusively installed
in customers homes. One option is to bury
them in the gateways customers use for their
entire suite of communications and enter-
tainment services. In this way, customers
need not even be aware femtocells are in the
house (or even know what femtocells are).
Nokia Siemens Networks has been marketing
just such a femtocell gateway following its
tie-up with RadioFrame Networks.
Welcome to your femtocell
The obvious alternative to swallowing all
the upfront costs is to pass some of them on
to customers. This, however, means convinc-
50
40
30
20
10
0
2007 2008 2009 2010 2011 2012
A
n
n
u
a
l

c
o
s
t

s
a
v
i
n
g
s

p
e
r

c
u
s
t
o
m
e
r

(
U
.
S
.

d
o
l
l
a
r
s
)
Source: Analysys Research (2007)
Annual cost savings per customer for a small operator
(with 5 million customers) deploying 3G femtocells
60% femtocell penetration 20% femtocell penetration
MART09-Feature.FINAL.indd 31 3/4/08 1:30:14 PM
Regardless, in markets like the UK,
where coverage is now patchy, an operator
plugging femtocells in this fashion might
not generate suf cient interest in them to
justify commercial deployment (according
to Parkin-Whites calculations). It may have
to dangle another carrot in the form of price
savings, and this represents a potentially
dangerous strategy. As soon as femtocells
become another weapon in the relentless
price war among mobile operators, they are
less likely to guarantee cost savings for the
operators themselves. The US$64m ques-
tion, as Bubley puts it, is whether an opera-
tor could knock down end-user prices and
still manage to reduce its own operating
costs enough to walk away with healthier
margins.
A related conundrum is whether price
cuts would trigger more indoor usage of
cellular services. Fixed-mobile substitution
has become an irreversible trend without
them. Last year, for example, Germanys
Deutsche Telekom lost 2.1 million fxed-line
phone customersan increase of 100,000
over the year beforepartly because
consumers switched to mobile (some may,
admittedly, also have changed providers
or turned to VoIP). In highly competitive
markets like the UK, where fat-rate mobile
tarifs now carry a big stash of free minutes,
many customers are already predisposed
to use their mobiles at great length when
indoors; they do not have to worry about
running up a huge bill.
Another problem in selling the concept is
that a more tech-savvy customer (the type
operators might assume are most recep-
tive to it) likely will fgure out he is, in efect,
doing his mobile service provider a favor by
shouldering its backhaul expenses. As noted,
operators save money with femtocells by
optimizing their network buildessentially
cutting back on outdoor equipmentand
using fxed broadband access, as opposed
to conventional cellular infrastructure, to
carry traf c. The customer paying the bill for
that broadband access might wonder if his
femtocell provider should help out.
Femtocells for data
Perhaps the biggest challenge for femto-
cells will be to spur the fxed-mobile substitu-
tion of data, as opposed to voice. While many
in the industry see this as their greatest op-
portunity, there are several reasons to doubt
they will have much impact.
First, why would a customer with
DSLa femtocell prerequisitewant to
use mobile data? It seems highly improb-
able he would choose to use a mobile
handset for Internet access with a PC in
the house (one must assume he has a PC
if he has broadband). Possibly, one might
argue, he wants to be untethered from
a fxed connection and be free to use a
laptop anywhere in the house. But such
a customer probably already would have
a router allowing him to connect to the
Internet via WiFi, and his laptop would
undoubtedly be WiFi-capable.
RadioFrames Keenan thinks one day the
number of laptops with embedded 3G chips
will outnumber those compatible with WiFi,
but right now few such devices are avail-
able. So the hypothetical customers laptop
would not work on a 3G network unless
he had bought a USB modem or dongle
from a service provider, in which case he
would also have had to sign up to a mobile
broadband contract. And if he took a mobile
broadband contract, why would he retain
the DSL line and pay for two broadband
services? Certainly in Austria and parts of
Eastern Europe, such customers are typically
mobile only.
In such circumstances, femtocell providers
may struggle to ofer their data customers
any cost savings. Fixed broadband has long
been available at fat rates, and 3G tarifs are
heading the same way. An integrated opera-
tor like France Telecom, with both fxed and
mobile networks, could attempt to bundle
DSL with 3G and ofer an all-in-one service.
But it would surely need to be at a massive
discount to the monthly price of DSL plus
3G to encourage take-up. Hoping femtocells
make that commercially feasible could be
expecting too much.
FMC
32

Telecommunications March 2008



www.telecommagazine.com


The mobile & wireless opportunity
The danger is mobile operators that fail
to adopt a large-scale [femtocell] approach
nd themselves expending great effort on
integrating a large number of 3G femtocells
without avoiding signicant investment in
outdoor networks
Andrew Parkin-White, Analysys Research
to 12Mbps, capable of high-quality video
download. STiMi is also designed for two-way
communications for VoD, interactive services
and authentication.
SARFT said its short-term goal is to make
CMMB available with terrestrial relays in 35
cities by summer, then expand to more than
320 cities after launching CMMB-Star, an
S-band satellite, in June. According to SARFT,
CMMB will ofer four to fve free national TV
channels, one to two local programs, and
charge for premium services such as VoD,
video games and special features.
Messy but clear
It is a messy picture and will take some
time to sort out, but the basic scheme seems
clear: 3G service becomes more attrac-
tive to potential customers when mobile
TV is added, as China Mobile is doing with
TD-SCDMA or as SARFT is ofering with the
separate CMMB service.
Both services will fnd their way into
the mainstream of mobile video com-
munications and entertainment, but
SARFT appears to have the upper hand in
feasibility and cost: CMMB ofers free TV
programming and the cost for premium
service can be as low as 20 yuan (US$2.7)
a month. Because it uses a lot of band-
width for transmission, the 3G version
of TV service will charge a fee for actual
data volume, running as high as 1,000
yuan (US$138) a month, which is exorbi-
tant for most customers. Also, because of
3G bandwidth constraints, mobile TV has
a limited ofering for live TV broadcast
and VoD.
Video content (e.g., phone, messaging,
entertainment) can be a strong catalyst for
3G success in China because many Chinese
use cell phones as personal entertainment
centers. The future of mobile TV lies in its
ability to attract a large number of cell phone
users. However, like 3G itself, entertainment
potential does not necessarily lead to opera-
tor proft.
With issues like regulation and unruly
competition, it remains to be seen how
the mobile TV market will spawn a new
ecosystem of content developers, service
integrators, handset manufacturers and
other value-added services. It will be a new
challenge for Chinas telecom and broadcast-
ing industries.
Lin Sun is a Beijing-based consultant with more
than 20 years experience in the Chinese tele-
com industry. lsun@chinanex.com
( China,continued from page 28)
MART09-Feature.FINAL.indd 32 3/4/08 1:30:24 PM
presents
See pioneering wireless M2M services and products, including wireless remote control and monitoring of xed
and mobile devices that are fueling advancements in enterprise and consumer markets.
Visit these Leading M2M Companies in One Pavilion at CTIA Wireless 2008
Get in the Zone!
Device connectivity
is here!
The M2M Zone Pavilion at CTIA Wireless 2008
Las Vegas, NV April 1-3
For full program details, visit http://www.ctiawireless.com/events/event_details.cfm?calID=674
or stop by Room N103 at the Las Vegas Convention Center
Presented by In Partnership with
M2M Carrier Strategies:
The New Rules
A Half Day Seminar at CTIA Wireless 2008
Tuesday, April 1, 2008 1:00 - 4:00 pm Room N103
Signature Sponsor
Smart wireless. Smart business.
Smart wireless. Smart business.
Session Sponsors
Sessions Include
Industry Research: Cellular M2M Services
Open Cellular Networks: Fact and Fiction
Evolution: From Vertical to Horizontal & Enterprise to Consumer
M2M Darwinism: Achieving Global Scale
p33M2M.indd 33 3/5/08 10:16:04 AM
The mobile & wireless opportunity Cellular backhaul
WHEN VOICE was the dominant
application, wireless backhaul was pretty
simple: When more capacity was needed,
providers simply provisioned another leased
T1 line from the local ILEC.
As wireless operators migrate to higher-
speed services (i.e., HSDPA and LTE), however,
throwing more T1 leased lines at the problem
is not an optimal solution from either a cost
or management perspective. According to
some industry estimates, deployment of new
3G and 3.5G wireless networks will require
two to four incremental T1s of traf c.
A study by ABI Research confrms wireless
backhaul, which currently accounts for only
US$14bn of a wireless operators capex
budget, will reach US$23bn by 2012. With
that in mind, wireless operators are looking at
ways to break the traditional T1 chokehold.
Take Sprint, one of the largest 3G and
WiMAX players in the United States. Although
T1 circuits remain its primary backhaul
solution, the operator is fnding other more
fexible mediums that lower its reliance on
traditional platforms.
The default traditional solution for 3G cell
site backhaul has certainly been LEC T1, says
Craig Cowden, vice president of cable opera-
tions and interconnected access solutions
at Sprint. I will say that for the last six to 12
months, we have been pursuing LEC substi-
tution alternatives. Certainly, Ethernet is the
dominant technical strategy, depending on
the physical medium, which could be either
copper or microwave.
All eyes on Ethernet
While Ethernet has long been the darling
of the LAN, and now increasingly of the
metro network, momentum for Ethernet-
based wireless backhaul continues to mount.
Ethernet/IP backhaul is arguably still a
nascent marketmaking up only 1 percent
of total mobile backhaul equipment sales
in 2006but it will account for about 41
34

Telecommunications March 2008



www.telecommagazine.com


Breaking
the
Wireless operators search
for new alternatives
by Sean Buckley
percent, or US$2.5bn, of worldwide backhaul
sales by 2010 (see Worldwide mobile backhaul
equipment revenue forecast).
Furthering the case for IP/Ethernet,
wireline standards eforts from the IP/MPLS
Forum and the Metro Ethernet Forum are cre-
ating a path operators can follow to migrate
toward IP/Ethernet (see IP/MPLS, MEF defne
blueprint for IP mobile backhaul, page 31).
Wireless operators and their wholesale
partners are responding. Sprint, for one,
takes a best-ft approach to wireless backhaul
that includes a mixture of traditional LEC T1
circuits, fber, microwave and even hybrid
fber coax. In its wireless backhaul network,
Sprint has a two-part strategy that accommo-
dates its 3G and WiMAX networks, but the 3G
group will leverage similar facilities if they are
on the same tower.
Sprints Cowden says a major focus
involves working with cable operators. Were
working with all the MSOs, some more ag-
gressively than others, to develop a DOCSIS-
over-Ethernet standard,he says. We want to
use the physical medium of their coax for Eth-
ernet, which is not just for cell site backhaul
but is also a huge drive for our enterprise last
mile access requirements.
Just the same, wholesale operators such
as Alabama-based, competitive wholesale
operator Southern Light acknowledge that
while most of their customers are still primarily
purchasing high-speed, TDM-based circuits, the
interest in Ethernet-based backhaul is growing.
You can fx a lot of problems in this
industry by throwing more bandwidth at it,
and obviously you get the most bandwidth
for your buck with Ethernet,says Eric Daniels,
Southern Light COO. Pure Metro Ethernet
[as opposed to Ethernet over SONET] is also
more scalable. We can take a customer from
10Mbps to 100Mbps to 1000Mbps in a few
minutes. Metro Ethernet requires much less
traf c engineering than TDM-based services,
so the whole process from service order to
delivery is that much more ef cient.
Living in the hybrid world
Even though an all-IP/Ethernet backhaul
network is certainly in the future, the near-
term reality is that wireless service is remains
primarily TDM-based.
$6
$5
$4
$3
$2
$1
$0
2006 2007 2008 2009 2010
R
e
v
e
n
u
e

(
U
S

b
i
l
l
i
o
n
s
)
Source: Infonetic Research
Worldwide mobile backhaul equipment
revenue forecast
ATM
Ethernet
Microwave radio
PDH NIU
SONET/SDH
Other
MART10-Feature.FINAL.indd 34 3/4/08 1:28:12 PM
One of the vexing issues with Ethernet is
network synchronization. Whereas TDM-
based networks have a predictable network
clock to ensure QoS for real-time services,
Ethernet lacks such a construct.
Jim Orr, principal network architect in
the wireless market development group at
Fujitsu, argues that a tight clock is required
for WiMAX, LTE and even current 3G wireless
networks. SONET, for example, delivers a
clock thats extremely tight, but Ethernet
does not because its a variable packet-based
system and in its native form it does not have
a stable network to run across,he says.
Emmy Johnson, principal of SkyLight Re-
search, agrees because even though wireless
operators are set on IP/Ethernet, they dont
want to compromise their cash cow: voice.
Ethernet obviously is a more economical
and ef cient way to carry data packets,she
explains. They want to start making that
migration, but the voice customers are their
bread and butter and they dont want to sac-
rifce many just to save the few coming up.
Currently, operators are considering two
near-term options:
Despite its promise of greater fexibility and bandwidth, proliferation of an all-
IP-based wireless backhaul network is arguably a work in progress.
Still, as wireless operators make their respective migrations from TDM-based
2G and 3G networks to an IP-based 4G WiMAX and long-term evolution (LTE)
future, wireless operators will need an IP backhaul network blueprint.
Enter the IP/MPLS Forums MPLS mobile backhaul initiative (MMBI) and the
Metro Ethernet Forums mobile backhaul initiative (MBH).
While various transitional technologies (e.g., pseudowires) that
accommodate TDM and IP are good near-term strategies, David Sinicrope, vice
chairman of the IP/MPLS Forum and chief scientist at Ericsson, argues a carrier
will need a forkliftwithout a prepared IP path.
Youll hear a lot of hype about pseudowires, but that only addresses 2G, 3G
and maybe 3.5G,he says. Where do you go from there? Many of these carriers
see long-term evolution as the ultimate goal. We want to create a framework
and technical specifcation to address whats here today. But how do you get to
tomorrow?
Building on the MMBI and MBH eforts, the two groups in conjunction
with the European Advanced Network Test Center showcased a public mobile
interoperability event during the MPLS and Ethernet World Congress and Mobile
World Congress in Barcelona last February. In April, IP/MPLS and MEF will
continue their work at the CTIA Wireless Show in Las Vegas.
The events were comprised of 15 vendors testing more than 85 devices
spanning microwave, base stations, testing, metro networking, core networking
and metro edge. Participants included Alcatel-Lucent, Cergaon Networks, Ciena,
Cisco, Ericsson, Harris Stratex Networks, Huawei, Ixia, Lightstorm Networks, MRV
Communications, Nokia Siemens Networks, Nortel, RAD Data Communications,
Spirent Communications and Telco Systems.
Touting their ability to work in harmony, the two forums demonstrated
how a wireless operator could migrate from todays ATM and TDM backhaul to
an Ethernet and IP/MPLS-based solution. In particular, these demonstrations
showed how traf c from a radio access network (RAN) could be carried over
Metro Ethernet and an IP/MPLS core. Additionally, the demo showed how
service providers can operate metro/aggregation networks based on provider
backbone bridging-transport engineering, transport MPLS and MPLS.
As interest in new mobile backhaul alternatives has grown among wireless
operators, both the MEF and IP/MPLS Forum have followed suit, albeit at
slightly diferent angles. MEFs MBH has created a service implementation
agreement for Carrier Ethernet services in mobile operators backhaul networks.
Upon completion, the IA will provide a simple standard specifcation for the
purchase and sale of Ethernet-based backhaul services.
Meanwhile, the IP/MPLS MMBI gives service providers a framework to
deploy MPLS and Ethernet to transport RAN backhaul traf c over access,
aggregation and core networks. In addition, MMBI can accommodate legacy, IP
or converged network confgurations.
Sinicrope points out that while both the MEF and the IP/MPLS Forum may
have started from diferent points, the work overall is benefcial to operators.
The work were doing is complementary to the services being defned
in the Metro Ethernet Forum,he said. They basically defne in their mobile
backhaul project [that] heres the interface to Carrier Ethernet, but they dont go
into any detail [about] how they are provided, and thats where we pick up.
IP/MPLS, MEF dene blueprint for IP mobile backhaul

www.telecommagazine.com

March 2008 Telecommunications

35

The mobile & wireless opportunity Cellular backhaul
Hybrid network. Embraced primarily
by European cellular operators, the hybrid
network strategy enables a wireless opera-
tor to maintain TDM for voice then run
higher-speed EVDO traf c over a packet-
based IP/Ethernet network. Emerging ven-
dors such as Exalt have built new products
that, for example, can run both TDM and
IP/Ethernet-based backhaul traf c in its
native form.
Pseudowires. With a pseudowire, a wire-
less operator can run TDM and IP traf c in
its native form over a common transport
network, an option adopted by European
operators such as Swisscom.
This has prompted a group of emerg-
ing vendors and alternative providers
to present solutions that can bridge
the gap between the TDM and Ethernet
worlds.
Telecom Transport Management, while
not ruling out the possibilities Ethernet could
bring to wireless backhaul, has developed
an underlying TDM infrastructure that can
carry either TDM or Ethernet depending on
customer needs.

Overall, Frank Mastrobattista, COO and


co-founder of TTM, says the ultimate goal is
fexibility.
The beautiful thing about microwave and
fber is you do get a lot of capacity, but you
also get fexible use of TDM or IP,he says.
Today, the vast majority of cellular backhaul
is TDM-based, and the majority of it will be
TDM for a while. As they migrate to 2G to 4G,
theyll keep the existing stuf on TDM and
deploy Ethernet for incremental capacity.
Even though it will accommodate various
backhaul modes, Sprint wont run two sepa-
rate backhaul networks.
Were not pursuing split backhaul solu-
tions where we would have totally diferent
backhaul for voice then potentially a separate
provider solution for data,Cowden says.
Right now, we are working with solutions
that, at the cell site itself, would have both an
Ethernet interface and a T1 interface, and the
egress port coming out of that box would be
fber. Within the fber itself, we would allocate
both T1 and Ethernet, and depending on the
service, it would ride either that T1 or Ether-
net within that fber.
MART10-Feature.FINAL.indd 35 3/4/08 1:28:23 PM
The mobile & wireless opportunity CDMA2000
THE CDMA MARKET is under
long-term pressure in Brazil and is being
swapped out for 3GSM in Australia. In
Venezuela, Mobilnet has announced plans
to overlay GSM on its CDMA network
to reach low-end subscribers. In India,
Reliance is migrating its CDMA network
to GSM in urban areas. Neither Verizon
nor Sprint has opted for long-term CDMA
migration strategies. Sprint is planning
a commercial launch of mobile WiMAX
in April, and last November Verizon
announced it will implement long-
term evolution (LTE), a 3GSM migratory
technology.
In the past CDMA has ofered superior
performance over GSM in terms of both
voice network capacity and data speeds.
Service providers accepted higher device
and infrastructure costs to beneft from the
performance capabilities CDMA ofered.
When introduced, CDMA 1X ofered
higher data rates and capacity relative to
the GSM/GPRS solutions ofered at the
time. And when GSM service providers
launched EDGE to achieve approximate
parity with CDMA 1X, the CDMA 1XEVDO
migratory generic became available. Again
the CDMA service providers could outper-
form their GSM-based competitors.
The performance gap with CDMA was
eliminated, however, with the introduction
of 3G-UMTS and HSDPA. Without a com-
petitive advantage in terms of perform-
ance, CDMA service providers have been
compelled to migrate toward 3GSM.
This ultimately renders CDMA an end-
of-life technology. However, its midterm
prospects are less certain. In particular:
Will CDMA growth in the United States
continue between 2008 and 2011, given
technologies like LTE will not be commer-
cially available until 2011 at the earliest?
And will growth in the combined markets
of the U.S., Korea, Japan and China ofset
declines seen in markets such as Brazil and
India, where CDMA is being phased out?
How long will CDMA prevail in the
marketplace, and what are the drivers and
inhibitors for continued market support?
Can CDMA compete with GSM at the
low end of the market?
Is there any future for the CDMA migra-
tory generics, namely CDMA2000 Rev B,
and UMB?
Even in the face of declining key market
sentiment, the number of CDMA subscrib-
ers will remain relatively fat between 2008
and 2011, supported by continued growth
in the U.S., India, Japan, Korea and China
(see Global CDMA subscribers forecast,
page 37).

36

Telecommunications March 2008



www.telecommagazine.com


In the face of waning support,
CDMA is now an end-of-life
technology that will see a long
twilight
by Phil Marshall
MART11-Feature.FINAL.indd 36 3/4/08 1:25:04 PM
U.S. stable
for medium term
In the U.S. market 51 percent of subscrib-
ers use CDMA technology, with Sprint and
Verizon being the dominant players. Yankee
Group forecasts have this proportional share
remaining relatively stable through to 2011.
The lions share of new subscribers in the U.S.
will come from the markets low end, which
is particularly price sensitive. Players such as
Motorola, Samsung, LG and Kyocera continue
to dominate the U.S. handset market. How-
ever, silicon providers such as VIA, in conjunc-
tion with their handset ODM partners, are
successfully penetrating the U.S. market with
devices optimized for low cost.
CDMA remains optimum technology
for voice communications. Players such as
Leap and MetroPCS are capitalizing on the
superior capacity of CDMA to address the
low-end market with fat-rate voice packages.
Both Leap and MetroPCS are benefting from
the increased availability of low-cost CDMA
devices.
Verizon and Sprint recently have launched
CDMA2000 DO Rev A to ofer broadband
data and VoIP solutions. Sprint initially plans
to use VoIP to replace its languishing push-to-
talk service, and Verizon is eager to migrate
its legacy mobile voice subscribers to VoIP.
While Verizon essentially has a network
designed for coverage at 850MHz, its CDMA
DO Rev A network has been launched at
segmentation strategy. KDDI has the minority
holding in a new venture recently awarded
a 2.5GHz WiMAX spectrum license. NTT
DoCoMo was not successful in procuring
WiMAX spectrum and is expected to deploy
LTE aggressively once it becomes available.
Yankee Groups expects KDDI to deploy
WiMAX aggressively, initially to complement
its CDMA network ofering then migrate from
CDMA to WiMAX as DoCoMo deploys LTE.
Yankee anticipates a similar strategy in Korea,
albeit more advanced as a consequence of
its early WiMAX initiatives under the guise
of WiBRO. In Latin America, China, and
India, Yankee expects service providers will
continue to use CDMA, particularly in rural
areas where it is well-suited to meet limited
mobility applications.
Even though CDMA is essentially an end-
of-life technology, Yankee Group expects it to
retain a robust market position in the short to
medium term, particularly in the U.S. market.
It will herald mobile VoIP in markets where
CDMA DO Rev A has been deployed and
continue to economically support traditional
voice services for low cost players such as
Leap and MetroPCS. Advanced technologies
such as WiMAX and LTE will erode its market
position after 2011, once they have reached
prime time as mobile solutions.
Phil Marshall is vice president of enabling tech-
nologies at Yankee Group
1900MHz. As reliable VoIP becomes available,
Verizon will aggressively refarm its 850MHz
spectrum for CDMA DO Rev A. This will vacate
PCS spectrum into which it ultimately will
expand LTE. Yankee Group does not expect
any U.S. service providers to launch CDMA
DO Rev B.
Sprint continues to invest in CDMA and
plans to launch WiMAX commercially under
the Xohm brand in April. The WiMAX initia-
tive is particularly speculative and, should
it remain under Sprints control, must be
coordinated more efectively with its CDMA-
based activities. Sprint supports in excess
of 30 mobile virtual network operators on
its CDMA network. While many of these
MVNO initiatives have failed, those targeting
new low-end market subscribers have seen
reasonable success.
KDDI opts for WiMAX
In Japan, KDDI has diferentiated itself
using CDMA with a high-focused, device-

www.telecommagazine.com

March 2008 Telecommunications

37

The mobile & wireless opportunity CDMA2000
Even in the face of declining key market
sentiment, the number of CDMA subscribers
will remain relatively at between 2008
and 2011, supported by continued
growth in the U.S., India, Japan, Korea
and China
M
i
l
l
i
o
n
s

o
f

s
u
b
s
c
r
i
b
e
r
s
Source: Yankee Group (2008)
Global CDMA subscriber forecast
(excludes impact of Movilnet Venezuela announcement
to overlay GSM on its CDMA network)

Asia Pacifc EMEA Latin America North America
0
50
100
150
200
250
300
350
400
450
500
2005 2006 2007 2008 2009 2010 2011

Will growth in the
combined markets
of the U.S., Korea,
Japan and China
offset declines seen
in markets such as
Brazil and India,
where CDMA is
being phased out?
MART11-Feature.FINAL.indd 37 3/4/08 1:25:13 PM
Green to go
for ber
FTTH Council Europe claims ber is
environmentally friendly
by Kendrick Struthers-Watson
Special focus FTTH
THE RESULTS of a unique research
exercise evaluating the sustainability and
environmental impact of fber networks
were announced at the FTTH Council Europe
event held in Paris last month.
Joeri Van Bogaert, president of the Council,
said, The results clearly demonstrate the
overall service and environmental benefts
of FTTH. The fndings stand as testament
that fber is a sustainable and future-proof
technology for the 21
st
century.
The research compared the environmental
impact of a typical FTTH network to its associ-
ated benefts and will provide the foundation
for further research on the subject by the
Council. Information and communication
technologies are providing new answers
for heavy energy consumption sectors such
as transportation and construction. FTTH
networks can bring environmental benefts
to many diferent areas and the Price Water-
house Coopers (PWC) study focuses specif-
cally on teleworking, telemedicine and home
assistance. Life Cycle Assessment was used to
generate the necessary quantitive data and
the work was conducted in accordance with
ISO 14040 standard.
This important study considered three
scenarios for the calculation: urban dense
areas, urban wide areas and rural areas. It
also covered four types of FTTH deploy-
ment: existing ducts or urban sewers,
traditional trenches, micro-trenches and
aerial hanging.
Maximizing the opportunity for new
services while minimizing the materials and
maintenance required, FTTH contributes to
reduced road travel, less transport infrastruc-
ture, and the introduction of innovative social
and government services. The study took into
account the total impact of network imple-
mentation and the full lifecycle of a typical f-
ber infrastructure from production of passive
equipment, transport, implementation of all
active equipment and power consumption
to end of life.
Christian Ollivry, chair of the SUDEFIB
Committee (the Councils Sustainable Devel-
opment and FTTH Committee) comments,
Broadband usage is an opportunity for
sustainable development and our mission
was to qualify the sustainable development
impact of FTTH solutions, looking at various
European cases. Beyond a 15-year timescale,
which is quite conservative, the research
shows that FTTH provides only positives for
Europe. The results are based on calculations
from teleworking and tele-assistance ap-
plications only, although FTTH solutions ofer
extensive social and economical benefts
over and above these, with the capability to
change societies for the better because of the
access to new services it brings.
38

Telecommunications March 2008



www.telecommagazine.com


MART13FEATURE.FINAL.indd 38 3/4/08 1:22:16 PM
Is your journey
really necessary?
From an environmental perspective,
fber brings with it a huge portfolio of
socio-economic benefts when translated
into teleworking, telemedicine and home
assistance. With all the relevant services over
fber, company personnel can work remotely
from the city-center of ce complex, thereby
relieving pressure on the demand for greater
of ce space and reducing the companys
annual rental and maintenance costs.
Of course, remote tele-workers can be de-
prived of human interaction with colleagues.
To counteract this, a small village community
was envisaged, for example, having a sepa-
rate of ce area where village teleworkers
would conduct their business activities for
say, three days a week. While each teleworker
would be working for a diferent company,
they would become of ce colleagues. An
additional upswing to this scenario is fewer
vehicle journeys resulting in reduced CO
2

emissions, less wear and tear on the road sys-
tem and an improved quality of life through
greatly reduced travel time each day.
From the current trends (2010 to 2011)
in FTTH networks, the report indicated that
up to 10 percent of the working population
could telework three days per week and 20
percent of the senior population (75 years
of age and over) could beneft from home
assistance.
Using INDATEs projection of FTTH users
(20 million for 2015), the research found that
for the frst 15 years of network implementa-
and less disruptive, henceforth reducing the
environmental impact and further increasing
the sustainability of fber networks over time.
Main outcomes
As a main quantitative fnding, the envi-
ronmental impact of the deployment of a
typical FTTH network will be positive within
less than 15 years on average considering
only the three selected services.
Further existing or developing applications
will emphasize these results.
Beyond their environmental-friendly as-
pects, FTTH solutions can ofer considerable
additional social and economical benefts.
If it is admitted we are now in the middle
of a new industrial era, FTTH solutions are a
key sustainable utility driver in this context.
Further research into the subject of
sustainable development is planned for
the near future. This could also include the
adoption of similar sustainable develop-
ment methodologies for a defned network
project by the FTTH Councils of APAC and
North America.

tion, greenhouse gas emission savings per


user were 330 kg, the equivalent of a car
traveling 2,000 kilometers. For the next 15
years, the savings are 780 kg eq. CO
2
or the
equivalent (e.q.) of the emissions caused by
a car traveling 4,600 kilometers. This is due to
the fact that the network is depreciated and
only part of the infrastructure needs to be
renewed.
Self-cleaning technology
Moving forward, the sustainability of FTTH
solutions will be increased signifcantly as
user experience grows and other felds not
assessed in this study such as supply-chain
management and energy demand are
considered.
Over the full network lifecycle, the use of
the network (power consumption) represents
only 6 percent of the total environmental
impact, while the production and deploy-
ment of the equipment totals more than 80
percent. However, with continuing innova-
tions taking place in the industry these
processes are becoming cheaper, quicker
FTTH
Green issues aside, the FTTH Council Europe is keen to win over the fnancial
community on more traditional business case arguments. And help was at hand
from Jakob Bluestone, a telecom analyst in the equity research department of
Merrill Lynch. Speaking at the FTTH Council Europes annual conference in Paris
last month, he said he took a more benign viewon the risks of high FTTH capex
from Tier 1 operators compared with the majority of analysts in the investment
community.
There is a lot of nervousness about high capex [projects],he admits. Telcos
havent got a good capex track record and there is an economic slowdown with
the prospect of higher infation this year.
Despite the prospect of a softening economy, at least in the short term,
Bluestone points out the telecom industry is still generally more attractive to
investors compared to many other industry sectors.
It is a hugely cash-generative sector and is reasonably resilient to economic
slowdowns,he says. [Telcos are concerned] about anything that could
undermine that, such as extensive fber access rollout, which requires a lot of
capital. There is a deep innate skepticism about high capex projects.
Bluestone used his presentation at the conference, however, to challenge
what he believes are commonly held bearish assumptions among fnancial
analysts on FTTH investment.
One of these, Bluestone says, is that IPTV will generate low returns (around
3 percent), which is of-putting for FTTH investors as it is much lower than
the cost of capital. Even if we accept that the IPTV return is low, there are still
other factors to consider,Bluestone says. FTTH can help operators keep their
customers and protect other revenue streams.
He also thinks analysts have overplayed the point that telcos cant take full
advantage of freeing up their real estate through fber investment in the last mile
(FTTH requires fewer local exchanges than the PSTN) as they have already sold
much of their real estate assets and now use them on a leaseback basis. It is
about 50/50 among telcos that have sold and not sold their real estate,he says.
Despite regulatory uncertainty and ongoing technological debate
surrounding FTTH in Europe, Bluestone is cautiously upbeat. It can make sense
to invest in FTTH, subject to being realistic about the returns that can be made,
he says.
Merrill Lynch bucks bearish consensus on FTTH investment
FTTH networks can bring environmental
benets to many different areas
specically teleworking, telemedicine and
home assistance

www.telecommagazine.com

March 2008 Telecommunications

39

Special focus
MART13FEATURE.FINAL.indd 39 3/5/08 8:59:15 AM
T
he prototypical dot.com business model
characterized by proftless growth
great top line results and terrible bottom
line resultsis giving way to much more
sophisticated IP services. These use subscriber
and application awareness to align precisely
the willingness to pay with the cost of service,
thereby creating services attractive to users
and proftable to providers. IP service control
technology makes this possible.
Service control requires knowledge of
what is going on in the network plus the abil-
ity to control it. Knowledge of a customers
applications and performance expectations
is also key to maximizing proft.
Many home of ce users, for example, are
involved simultaneously in sending and re-
ceiving e-mail, while using the same broad-
band connection for IP telephony. A voice
call is extremely sensitive to packet jitter and
delay, while e-mail delivery is sensitive to the
time it takes to deliver the entire message.
A broadband connection, therefore, that
minimizes delay and jitter for the voice call
and allocates the remaining bandwidth to e-
mail will provide better user satisfaction (and
a higher willingness to pay) than a broad-
band connection that lacks this application
awareness.
In general, service control involves knowl-
edge of how subscribers and applications use
the network, thereby enabling a provider to
develop and enforce policies that balance the
user experience with the cost of network re-
sources. This allows providers to manage and
charge for premium Internet service instead
of simply selling bandwidth and to act as a
wholesale provider of content-based services.
Protect & defend
A second element of IP service control is
protecting the network and its users from
the negative impact of rogue applications
and attacks. This is essential to maintain-
ing pricing premiums. Customers must be
assured of data integrity and security; service
providers must protect themselves from
theft of service and service disruption.
Three of the most important network pro-
tection requirements are distributed denial of
service mitigation, peer-to-peer traf c control,
and domain name service protection.
DDoS mitigation is especially important
to large enterprises with highly recognizable
brands, because DDoS can disrupt the entire
operation. Peer-to-peer traf c, typically used
for fle sharing, can result in a small number of
users consuming a disproportionate amount
of total network capacity. Sophisticated
network operators do not block this traf c
arbitrarily but use network intelligence to
manage the traf c to maximize their revenue
and satisfy legitimate user needs. DNS is es-
sential to maintaining a continuous Internet
presence. This of course makes it an attractive
hacker target and requires protection.
IP service control is achieved using deep
packet inspection (DPI).
Service provider access and aggregation
infrastructure (e.g., DSLAMs, OLTs, switches,
routers, BRAS, MPLS) is based on Layer 2,
Layer 3 and some Layer 4 technologies. This
infrastructure has limited capabilities to iden-
tify specifc customers, Internet destinations,
applications and services.
DPI technology, in contrast, extends from
Layer 2 through Layer 7 including applica-
tions. DPI can go beyond the basic service
provider access infrastructure to monitor,
shape, direct and shutdown traf c. It also
can identify and classify users, applications,
sources and destinations. This permits inte-
grated content control consisting of service,
security and transport regulation.
Service control includes analysis of
packets at Layer 2 through Layer 7, control
of peer-to-peer traf c, and the capability of
ofering tiered services. Security includes
state and fow tracking, in-line fltering and
blocking, signature detection, and protocol
and traf c anomaly detection. Transport
control includes encapsulation, switching
and routing functions.
Converging trends
With the beneft of hindsight it may seem
obvious these capabilities are required to
ofer high value-added and premium-priced
IP services proftably. However, a number of
trends needed to converge before this can
be achieved.
First it was only recently that the underly-
ing device technology became capable of
performing DPI at wire speedwithout
introducing unacceptable packet transport
latency. Second, providers needed an under-
standing of which IP services users would be
willing to pay. Did anyone really know you
could make money selling ring tones?
Finally, security threats evolved with the
Internet. DDoS and DNS attacks are just as
creative and innovative in their own perverse
way as are more socially acceptable innova-
tions such as video streaming and VoIP.
IP service control technology helps deliver
innovative services proftability and develops
the trust needed for content providers, IP
service providers, fnancial intermediaries
and the general public to work together to
create useful, entertaining new services.
Michael Kennedy is co-founder and managing
partner of Network Strategy Partners
(mkennedy@nspllc.com)
IP service control goes
beyond switching &
routing
DDoS and DNS attacks are just as creative
and innovative as are more socially
acceptable innovations such as video
streaming and VoIP
TechSpeak Michael Kennedy
40

Telecommunications March 2008



www.telecommagazine.com


MART15-TechSpeak.FINAL.indd 40 3/4/08 1:20:24 PM
Presented by: In Partnership with:
This 1-day conference explores the latest technologies, products, business cases, and
early mobile operator experiences with next generation solutions, as well as long term
operator strategies for mobile backhaul.
9:00-9:30am Introduction: Re-Architecting Backhaul for Next Gen: 3G, 4G and Beyond
Speaker: Michael Howard, Principal Analyst & Co-Founder, Infonetics Research
9:30-10:45am All-IP Mobile Backhaul Business Cases and Timeframes
Moderator: Michael Howard, Principal Analyst & Co-Founder, Infonetics Research
11:00-12:15pm Using Ethernet to Backhaul Mobile Voice with Clock Synchronization
Moderator: Michael Howard, Principal Analyst & Co-Founder, Infonetics Research
12:30-1:45pm Scaling for the Future: Backhauling Mobile WiMAX
Moderator: Richard Webb, Directing Analyst, WiMAX, WiFi, and Mobile Devices, Infonetics Research
2:00-3:15pm Converging 2G/3G and UMTS/CDMA on Common Backhaul Infrastructure
Moderator: Stphane Tral, Principal Analyst, Service Provider VoIP, IMS, and Mobile Infrastructure, Infonetics Research
3:30-4:45pm Ensuring Video Quality in Backhaul Networks
Moderator: Jeff Heynen, Directing Analyst, IPTV and Next Gen Oss/BSS, Infonetics Research
For more information and to register, visit www.telecommagazine.com/2008/CTIA
p41TCSBackhaul.indd 41 3/4/08 2:37:53 PM

Mobile TV hampered by high
prices and low reliability
THERE IS a fast-growing army of ex-mobile TV users in the United States and Europes
major mobile markets. Disgruntled by high prices and dissatisfed by the services poor
quality and reliability, consumers are turning their backs on mobile TV. So much so, in
fact, that the number of ex-mobile TV users now outnumber mobile TV users in Germany,
Spain, France and Italy by more than two to one.
This is the key fnding of a survey conducted by M:Metrics, the results of which were revealed
during Mobile World Congress held in Barcelona last month. By comparing mobile TV usage
patterns in January 2007 with those during November 2007 in the UK, Germany, Spain, France,
Italy and the U.S.more than 34,000 mobile users surveyed in totalM:Metrics found the size
of the mobile TV audience had increased by 36 percent. The bad news for mobile operators,
however, is that the number of former mobile TV users had grown by a whopping 68 percent.
Price was the biggest turn-of for ex-users,says Paul Goode, senior analyst with M:Metrics,
as over half of the ex-user group cited this as the main reason for fnishing with the service.
The second most popular reason for mobile TV disillusionment among M:Metricssurvey re-
spondents was poor quality and reliability, cited by about 25 percent of ex-mobile TV users across
the countries surveyed (and ranging from 22 percent in the U.S. to nearly 30 percent in Spain).
As the M:Metrics mobile TV survey is commissioned by Tellabs, a U.S.-based supplier of trans-
port kit aimed at increasing the quality and reliability of data services in a cost-efective way, the
survey explored this issue a little deeper compared with a similar survey undertaken by
M:Metrics last year (also commissioned by Tellabs).
We asked those who cited quality and reliability as a reason for not using mobile TV whether
they would be more likely or less likely to use the service again if these issues were resolved,
Goode explains. Of this group, 60 percent said they would most likely watch mobile TV again.
And if we include those who had never watched mobile TV but still cited quality and reliability
as a reason for not using the service, we end up with a total value of US$270m of lost revenue
across the markets surveyed.
The US$270m fgure is arrived at by assuming 932,000 ex-mobile TV users (put of by poor
service quality) and an additional 3.2 million frst-time users (held back by quality issues) start
using mobile TV as a consequence of the quality and reliability issues being cleared up. It also
assumes half of this total, around 2 million, will subscribe for a year to a mobile TV package with
an average monthly fee of US$11.30. This would generate the estimated US$270m revenue
fgure and more than double the size of todays mobile TV audience in the countries surveyed.
If the industry addressed the price opportunity, Im sure the growth opportunity for mobile
TV is larger still,Goode adds.
Ken Wieland
42

Telecommunications March 2008



www.telecommagazine.com


Endpoint M:Metrics
IPTV providers keen on
monitoring quality
Results of an MRG study confrm that
video quality is critical to the growth and
success of IPTV, signifcantly afecting as it
does customer support calls and customer
churn. The study also revealed that IPTV
service providers feel their existing moni-
toring solutions lack an accurate measure-
ment of perceptual end-user quality. Some
of the key fndings are: 84 percent of IPTV
service provider respondents reported
video quality monitoring as critical or a
very important part of their video initiative;
more than 90 percent of respondents were
notifed about service quality problems
from subscriber calls; 77 percent say video
quality is a main reason for customer
churn; and less than half of respondents
conduct deep packet inspection.
Mobile voice
to eclipse xed
Mobile voice calls will exceed those on
fxed networks in Western Europe by the
middle of 2008, according to new re-
search from Analysys. The latest Telecoms
Market Matrix from Analysys Research
shows that in the UK, where patterns of
consumption are close to the European
average, mobile voice usage should over-
take fxed voice in the second quarter of
2008. Dave Millett, operations director at
VoIP provider Inclarity, comments: Its no
surprise mobile voice calls are expected to
exceed those on fxed networks. Lets face
it, mobiles are small and convenient for
people who are constantly on the move,
even more so with a rapidly changing
work culture which is moving toward fex-
ibility, home and remote working.
Poor mobile voice quality
According to study results revealed by
Ditech Networks, nearly 40 percent of
mobile calls fall below industry minimum
standards for voice quality. The report,
which analyzed 630 million calls from 16
mobile carriers in 12 countries, revealed
that in mature markets such as the United
States and Western Europe, 23 percent
of calls fall below industry minimum. In
rapid growth markets like the Middle East,
India and Latin America, 59 percent of
calls fall below industry minimum. It was
also found that ambient noise, or noise
originating in the callers environment
and entering the devices microphone,
was rated objectionableon up to 50
percent of all calls in some regions.
PRIME NUMBERS
GERMANY SPAIN FRANCE ITALY UK U.S.
Number of mobile phone
subscribers watching mobile TV
(3 month average Nov. 07) 281,416 305,459 420,629 497,521 482,900 1,737,004
Number of ex-users for every
current user (3 month average
Nov. 07) 2.1 2.4 2.7 2.2 1.8 1.3
Growth in audience for mobile
TV (3 month average Jan. 07
to Nov. 07) 51% 17% 44% 23% 25% 45%
Growth in ex-users of mobile
TV (3 month average Jan. 07
to Nov. 07) 159% 74% 147% 37% 55% 49%
Quality & reliability cited by %
of ex-users as reason for not
watching (Oct. 07) 26.7% 28.4% 25.4% 22.7% 30.0% 22.0%
Source: M:Metrics (2008)
The effect of poor quality and reliability on mobile TV
MtT17 Endpoint.FINAL.indd 42 3/4/08 1:17:42 PM
CTIA WIRELESS 2008
The convergence of more than 1,100 exhibiting
companies, dozens of industries and over 40,000
professionals from 125 countries all working toward
the common goal of revolutionizing wireless.
Whether in broadband convergence, enterprise,
advertising, social networking or entertainment
CTIA WIRELESS 2008 is the global marketplace
connecting wireless and you.
April 1-3, 2008
Las Vegas, NV, USA Las Vegas Convention Center
www.ctia.org/ctiawireless
The most important global technology event of the year!
C3CTIA 0308.indd C3 2/28/08 2:39:38 PM
2008. Verizon. All Rights Reserved. All services are provided by Verizon in accordance with certain rates, terms, conditions, and restrictions set forth in Verizons taris or in applicable agreements between Verizon and the carrier.
Talking about the elephants in the room is one way
Verizon Partner Solutions helps our partners succeed.
Thats why we recently convened a customer summit to
hear your beefs and to troubleshoot solutions. You talk.
We listen. For after all, when you succeed, we succeed.
1
-
888
-
776
-
9411
verizon.com/partnersolutions/portfolio
High time we talked about it.
C4Verizon Mar08.indd C4 2/28/08 2:41:01 PM