umts -wlan roaming in hot spot areas a techno -economic · pdf fileumts -wlan roaming in hot...

UMTS-WLAN Technoeconomics IST

UMTSUMTS--WLAN roaming in hot spot areasWLAN roaming in hot spot areasA technoA techno--economic studyeconomic study

Anne Cerboni – France Télécom R&D

Jarmo Harno, Ilari Welling – Nokia Research Center

Dimitris Katsianis, Dimitris Varoutas – University of Athens

Kjell-Ove Kalhagen - Telenor

3rd IEEE Workshop on 3rd IEEE Workshop on WirelessWireless Local Local AreaArea NetworksNetworksSeptember September 2727--28, 2001, Newton, Massachusetts28, 2001, Newton, Massachusetts


Work conducted withinWork conducted withinEuropeanEuropean R&D projetR&D projet

wObjective : Techno-economic evaluations of broadband fixed and mobile solutions for telecommunications services

wPartners: Telenor, Nokia, France Télécom, Deutsche Telekom, Universidade de Aveiro, University of Athens, Atlantide


What is seamless What is seamless mobile IP service provision?mobile IP service provision?

wMultimedia services offered via a 3G mobile network complemented with WLANs(Hiperlan/2) in hot spots; both networks belong to the same operator.

wThe access networks ensure network discovery, seamless handover, qualityof service requirements, and transparent roaming.

wThe core network is IP.

wArchitecture is that studied in the IST/MIND project.


UMTS UMTS –– WLAN business caseWLAN business caseQuestions Questions andand challengeschallenges

w Question: Is UMTS+Hiperlan2 worth the investment? Under whatconditions?w Goal: to construct a techno-economic business case simulatorw Demand forecasting

– Uncertainty in market uptake in a highly competitive context

– Many services are currently unknown

w Dimensioning rules– Technical understanding of the target architecture and upgrades

– Ability to model the architecture

w Pricing– Reference prices not available for broadband mobile services

– New business models: advertising, 3rd-party pays, e-commerce commissions, volume or session based pricing

w Confidentiality– market data, costs, roll-out strategies are manufacturer and operator secrets


TheThe business case business case approachapproach: TONIC : TONIC tooltool

NPVNPV IRRIRR PaybackPeriod

PaybackPeriod

Economic Inputs

Cash flows, Profit & loss accounts

Geometric Model

Services Architectures

Year 0 Year 1 Year n Year m. . .

Demand for the Telecommunications Services

DB

RevenuesOA&MCosts

Life Cycle Cost

Life Cycle Cost

First Installed Cost

First Installed CostInvestments

Risk Assessment


3G 3G and and 3.5G 3.5G demand curvesdemand curves

Mobile subscriber penetration:3G, 3.5G and the sum

0,0

10,0

20,0

30,0

40,0

50,0

60,0

70,0

80,0

2002 2003 2004 2005 2006 2007 2008 2009 2010

3G + 3.5G

3G

3.5G


Service classesService classesCircuit/packet

Switched Bandwidth

class Quality of

Service class Sample services Nominal data

rate (kbps)* Supporting

network Circuit Narrowband Conversational voice call 33 UMTS

Circuit W ideband Conversational

video call, enhanced m-commerce

390 UMTS

Packet Narrowband Conversational voice over IP 33 UMTS

Packet W ideband Conversational video call, games 230 UMTS

Packet Broadband Conversational video conf. 1227 WLAN

Packet Narrowband Streaming

on rich call (e.g.audio clips)

33 UMTS

Packet W ideband Streaming

rich call (incl. video clips)

230 UMTS

Packet Broadband Streaming

Near video-on-demand

1227 WLAN

Packet Narrowband Int_backgr short msg., WAP 3.4 UMTS

Packet W ideband Int_backgr E-mail, Internet 16.4 UMTS

Packet Broadband Int_backgr

Large file transfer, appl.

200 WLAN

*overhead included


From From service classes to network service classes to network dimensioningdimensioning

11 classesBusinessSegment

11 classesResidentialSegment

mn/dayResid.

mn/dayBusiness

Daily usagefor each service class*

20%

30%

% of usage duringBusy hour Nominal bandwidth x

mn of use during BH

Number of businesscustomers each year

per service class*

Number of resid.customers each year

per service class*

BH kbpsfor eachService

class

required BH network capacity*time-dependent parameters


Network Network Investments Investments (1)(1)GSM/GPRS to UMTS: new radio access network, some re-use of NSSHiperlan/2: Radio access points & controllers, routers and serversInterworking (architecture developed within European IST/MIND project)

UMTS CN

GGSN / FASGSN

HLR

SGSN

UTRAN

RNCMN

AccessServer / HA

Backbone

AAA

BAN

AAA

BARMN ( IWU)B M G

UTRAN

Corporate

AccessServer/HA

BAN

UMTSPS CN

S G S N GGSN/FA

S G S N

I W UBARM N

RNCM N

HLR

AAA

B M G


Network Network InvestmentsInvestments (2)(2)

wUMTS

– Radio Access Network (RAN): Base stations (Node B) and radio network controllers (RNC)

– Network Subsystem: SGSN, GGSN, HLR/AuC, MSC upgrades, MSS, MGW, CPS, HSS

wWLAN

– Access Points, Access Point Controllers, Access Routers, Authentication server, Inter Working Unit

wCommon equipment

– Billing and customer care, Service platforms (service implementation, m-commerce, location servers, streaming servers, rich messaging servers, etc.)


Initial Initial resultsresults

w Small country– With / Without WLAN

• NPV = 948 / 859 M€

• IRR = 38% / 37%

• Payback = 5.6 / 5.5 years

w Large country– With / Without WLAN

• NPV = 5597 / 4413 M€

• IRR = 24% / 22%

• Payback = 7.1 / 7.3 years


Initial Initial Economic resultsEconomic results

Cash Ba lance

-10.000

-5.000

0

5.000

10.000

15.000

20.000

2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Years

ME

uro

Cash Balance LY

Cash Balance SY

Cash Balance LN

Cash Balance SN


Sensitivity analysisSensitivity analysis

wWhat if…?

wApproach

– Select critical input parameters

– Set boundaries for their variation

wResults

– Impact on Net Present Value

– Impact on Internal Rate of Return


Large country NPV Large country NPV sensitivitysensitivity

-4 000

-2 000

0

2 000

4 000

6 000

8 000

10 000

12 000

14 000

Low Va lue -2 761 -1 252 740 8 869 3 608 6 578 6 381 6 010 5 848 5 597 5 600 5 599 5 599 5 598

High Value 13 949 10 455 2 324 7 586 4 615 4 813 5 184 5 346 5 676 5 594 5 595 5 595 5 596

Tariff

Level

Total

Penetrati

on

End

Market

Share

OA M

Start

Market

Share

UMTS

BTS Cost

Terminal

Subsidy

Marketin

gChurn

WLAN

Building

Begin

WLAN

A P Contr

WLAN

Brain AR

WLAN

A P

WLAN

A P Inst


Large country IRR Large country IRR sensitivitysensitivity

-5%

0%

5%

10%

15%

20%

25%

30%

35%

40%

Low Value 0% 6% 12% 32% 19% 28% 26% 25% 24% 24% 24% 24% 24% 24%

High Value 38% 31% 16% 28% 20% 22% 22% 23% 24% 24% 24% 24% 24%

Tariff

Level

Total

Penetrati

on

End

Market

Share

OAM

Start

Market

Share

UMTS

BTS Cost

Terminal

Subsidy

Marketin

gChurn

WLAN

Building

Begin

WLAN

AP Contr

WLAN

Brain AR

WLAN

AP

WLAN

AP Inst


Initial conclusionsInitial conclusionswWLAN improves the NPV by 27% in the large country and by 10%

in the small country.

wWLAN represents less than 5% of the whole investment cost, but contributes 9% of total revenues.

w Tariffs, market demand, and market share are shown to be the most critical factors for success.

w Further work will concentrate on capacity-driven dimensioning and more limited (eg campus-wide) WLAN deployment scenarios.

w For more information, see websites

http://www-nrc.nokia.com/tonic/

http://ist-mind.org

umts -wlan roaming in hot spot areas a techno -economic · pdf fileumts -wlan roaming in hot...

Documents