[PPT] - Techno-economics for next generation fixed access networks PowerPoint Presentation

SLIDE 1

Author: Thomas Monath, T-Systems Enterprise Services GmbH Co-Author’s:

Sandro Krauss, Mario Kind, T-Systems Enterprise Services; Chris Hawinkel, Jan Van den Abeele, Alcatel; Antonio Elizondo, TID; Antoine Chuquet, Yann Denis, FT; Damiano Raspollini, TI; Daniel Ågren, Acreo; José Miguel Santos, Ricardo Afonso, PTI; Panagiotis Saltsidis, Thomas Kallstenius, Ericsson;

«Techno-economics for next generation fixed access networks »

Methodology and framework – some key assumptions; Major results of the fixed broadband access evolution scenarios

SLIDE 2

2

MUSE “Gaming field” Objectives & Research Questions of MUSE techno-economics Methodology & Framework Assumptions Use Case Definitions and major Results

Migration use cases Native Ethernet cases

Main Conclusion of the fixed broadband access evolution

scenarios

Outline

SLIDE 3

3 DSL Kerb/Cabinet Access multiplexer Edge node FTTH Access Aggregation Network Wireless feeder Application server Subscriber, QoS, and OAM management Internet PSTN Home gateways Application servers DSL Kerb/Cabinet Access multiplexer Edge node FTTH Access Aggregation Network Wireless feeder Application server Subscriber, QoS, and OAM management Internet PSTN Home gateways Application servers

Lower cost Increased intelligence in access Functionality moving closer to subscribers Fixed-mobile convergence New multimedia applications Applications initiated from various CP devices High bandwidth Increased fibre penetration Automated OAM New revenues by “Multi-service” access

MUSE “Gaming field” - Network evolution towards NG multi service capability

Multi-service capable NGN leads to requirements:

> Platform functions converge > Service & Network platform elements require new functions to be implemented

(NT … aggregation nodes … BRAS )

> Bottleneck in the last mile must be overcome

SLIDE 4

4

Objective & Research Questions

Provide techno-economic validations of MUSE architectural and functional choices:

> Two major technology trends in NG Access and Aggregation

using Ethernet transport technology between home gateway and service edge in order to provide IP based services:

> Pure Ethernet based switching technology (IEEE 802.1 xx) based on

MAC in combination with service (S)VLAN “pipes” and

> “IP” based network architecture using IP forwarding functionality

based on the IP packet source and destination information

> Functional evolutions:

> PPP DHCP, QoS, Multicast

Approach:

> Investigate several network migration and “Greenfield” scenarios

(actually focused on Network Access Provider)

SLIDE 5

5

Value-Chain Players & Business Roles

Techno-economic validation based on Network Access Provider (NAP)
!

! ! !

"

" " "

#!$

SLIDE 6

6

Methodology – Cash Flow Analysis

NPV NPV IRR IRR Payback Discounted cumulated cash flows Cash flows, Profit & loss accounts

Year 0 Year 1 Year n Year m . . .

Demand for the Telecommunications Services

Revenues OA&M Costs First Installed Cost First Installed Cost - CAPEX Investments

IST-2000-25172 “TONIC”

Economic Inputs Services Architectures

Tool - environment

Geometric Model DB …

SLIDE 7

7

General Assumptions

> Case study approach with focus on Access & Aggregation Network > Study period 2005 – 2010 > Business Model seeds back a certain percentage of the revenues to all

players (NAP share is considered only, 100% market share applied)

> Discount Ratio equals to 10% for all scenarios > Network elements in general will be modelled based on publicly

available price information combined with learning curves & maintenance costs assumptions

> OA cost model considers:

Network Operations incl. Management, Control,..
Employees depending expenditures
Provisioning costs
Sales and Marketing

SLIDE 8

8

General Assumptions –

Market Forecast and related Service Classes

Service Class Downstr. Capacity Tmax/ down [Mbps] Upstr. Capacity Tmax/ up [Mbps] Basic- Residential 1 0,5 Silver- Residential 3 1 Gold- Residential 10 2 Premium- Residential 50 10 Platinum- Residential 100 100 Basic- Business 2 2 Silver- Business 10 10 Gold- Business 50 50 Premium- Business 100 100 Platinum- Business 1000 1000

Service class definition:

Penetration evolution based on various consultancy reports

Sources: Ovum, Yankee BB market Europe, OECD, IST, survey between the MUSE partners

MUSE penetration potential forecast

based on service bottom up model

0,00% 10,00% 20,00% 30,00% 40,00% 50,00% 60,00% 2005 2006 2007 2008 2009 2010

Platinum-Business Premium-Business Gold-Business Silver-Business Basic-Business Platinum-Residential Premium- Residential Gold- Residential Silver- Residential Basic-Residential

Example - Germany T-Com - Dec 2005: about 8 Mio DSL, about 21% of the total fixed lines; Market Forecast Service bottom up model

SLIDE 9

9

General Assumptions

Area Description and Potential Market

Local Exchange Central Exchange Dense Urban: 2 x 32,768 Urban: 4 x 16,384 Suburban: 8 x 8,192 Rural 32 x 2,048 Low Dens. Rural 32 x 2,048 Total Connectivity: 65,536

LL1 LL2 LL3 LL4 LL5 LL7 LL6 FP5 FP4 FP3 FP2 FP1 FP0 FP7 FP6

Customer

Customer- LEx Customer- Cabinet Cabinet- LEx per Building per km² Residential % Business % Dense Urban 1.267 207 1.060 2 128 512 64 7.187 70 30 Urban 1.417 290 1.127 4 64 512 32 3.116 76 24 Suburban 1.750 467 1.283 8 64 2.048 4 1.043 82 18 Rural 2.033 683 1.350 32 32 1.024 2 177 88 12 Low Density Rural 2.767 1.217 1.550 32 32 2.048 1 8 94 6 Potential Customers Building s per LEx Cabine ts per LEx Geo-Model (Total Europe) 65.536 Total Potential Customers Covered LEx Average link lengths [m]

Input of involved network operators

SLIDE 10

10

General Assumptions –

Service based Revenue Model – Network Access Provider Service Revenue Model based on consultancy reports and benchmarks

Usage charge Service charge Access charge Connection fee

Business Residential

AoD, VoD, VoD SDTV, VoD HDTV AoD, Content Storage, Sensors/ Monitoring Broadcast Video, Broadcast SDTV, Broadcast HDTV, Gaming, Monitoring VoIP,HSI , Broadcast audio, Conferencing VoIP , Broadcast audio, BSI, HSI, Conferencing

Usage charge Service charge Access charge Connection fee

Business Residential Pricing scheme

AoD, VoD, VoD SDTV, VoD HDTV AoD, Content Storage, Sensors/ Monitoring Broadcast Video, Broadcast SDTV, Broadcast HDTV, Gaming, Monitoring VoIP,HSI , Broadcast audio, Conferencing VoIP , Broadcast audio, BSI, HSI, Conferencing

0,00 50,00 100,00 150,00 200,00 250,00 2005 2006 2007 2008 2009 2010 Basic-Residential Silver-Residential Gold-Residential Premium-Residential Basic-Business Silver-Business Gold-Business Premium-Business

Monthly Tariff - 5% / Year

Related to connectivity classes Related to connectivity classes Basic service Basic Silver Gold Premium Platinum Residential market 180 360 600 900 1 200 Business market 480 840 1 440 2 400 4 800

NAP: Default Input for residential and business market:

SLIDE 11

11

General Assumptions –

Network Element Model

Learning curve

5 10 15 20 25 30 35 40 45 50 2003 2004 2005 2006 2007 2008 2009 2010 Ether_CE_Swith Basic

[k ]

Network element: Switch Router GbE Application: Central Exchange LEx LEx Core

Central Exchange (CEx)

Downlink interfaces Uplink interfaces Downlink interfaces Uplink interfaces 16 slots Basic Equipment (Chassis +Basic modules +Software) 1 OC3c-POS-Interface (longdistance; SM) 4 1 OC3c-POS-Interface (longdistance; SM) 4 1 1000Base

LX

(mediumdistance; SM) 8 1 1000Base

LX

(mediumdistance; SM) 8 1 1000Base

LX

(longdistance; SM) 8 1 1000Base

LX

(longdistance; SM) 8 1 1000Base-SX (short distance; MM) 8 1 1000Base-SX (short distance; MM) 8

Switch Router GbE

Network element model based publicly available price information combined with learning curves & maintenance costs assumptions

SLIDE 12

12

Use Case Definition –

Migration Scenarios

today

BE QoS

ATM Ethernet PPP DHCP

BE QoS
%&

%& %& %& ' ' ' '

ATM Ethernet

PPP DHCP L2AM L3AM

%(

%( %( %(

L2AM L3AM

) ) ) )

Functionality moving closer to subscribers Layer 3 AM (IP) L2 Access Mux Lower Investment Ethernet transport ATM transport Multi Service QoS classes Best Effort

Driver Trend Today

SLIDE 13

13

Ethernet AN IP (v04/06) AN

Use Case Definition –

Migration Use Cases - Overview

FP1

In-house cabling

FP5

Local Exchange

FP7

Central Exchange

FP3

Cabinet Location

FP2

Bilding

FP0

CPE

FP4

Branch (1)

FP6

Branch (2)

First Mile Aggregation IP- Backbone %* %* %* %*

!

"# $%

(A-B)

Switch AN

STM4 STM4

DSL NT BRAS

(A-C)

Switch AN

GbE GbE

DSL NT BRAS

(A-D)

Switch AN

GbE GbE

DSL NT BRAS

AAA Fct. ATM Interf. QoS; CoS Multicast New xDSL

All cases combined with First Mile remote

DSLAM solution!

FTTC

Remote unit & '(&%

GbE

GbE GbE

still ATM but new services and functions

network element invests

input from the MUSE consortium vendors

SLIDE 14

14

10,0 Mio
Migration to a QoS enabled multi-service architecture based on Ethernet (A-C) or IP

forwarding (A-D) are more economic than the ATM-based evolution scenario (A-B)

Similar CAPEX for Ethernet and IP base architecture
5,0 Mio

0 Mio 5 Mio 10 Mio 15 Mio

2005 2006 2007 2008 2009 2010

Comparison: Cash Balance - Migration

Fibre to the Cabinet (FTTCab) – up to 50 Mbit/s Dense Urban area (about 30.000 customers connected in 2010)

A Reference A-B A-C A-D

(1) Migration from a best effort access network (A) to a QoS enabled multi- service architecture based on Ethernet (A-C) or IP forwarding (A-D) are more economic than the ATM-based evolution scenario (A-B)

+,- +,- +,- +,-

Major Results – Migration Use Cases

+&- +&- +&- +&-

(2) Almost the same investments for Pure Ethernet based switching technology (IEEE 802.1 xx) or “IP” based network architecture using IP forwarding functionality

+(- +(- +(- +(-

(3) In comparison to the other migration

ptions - it is not a choice to stay with

the current service offer (1-3/6 Mbit/s) as an incumbent network operator

SLIDE 15

15

Major Results – Migration Use Cases

Update the network with new products based on ADSL 2+ is more economical in

combination with Ethernet based access and aggregation network.

DSL upgrade more economical with Ethernet Access and Aggregation
10,0 Mio
5,0 Mio

0 Mio 5 Mio 10 Mio 15 Mio

2005 2006 2007 2008 2009 2010

Comparison: Cash Balance - Migration

Fibre to the Local Exchange Only (FTTLEx) – up to 16 Mbit/s Dense Urban area (about 20.000 customers connected in 2010)

Reference A-B ADSL 2+ A-C ADSL 2+

(1) Migration scenario A-B FTT(LEx) - update with new products based on ADSL 2+ still based on ATM aggregation network but including new functionalities (QoS, Multicast) increases the Reference case

(1) (2)

(2) The case is more profitable in combination with an Ethernet aggregation network

SLIDE 16

16

Major Results – Migration Use Cases)

(1) Multicast saves network resources (2) QoS enables services and possibly new revenue streams (3) Dynamic Host Control Protocol (DHCP) as IP auto-configuration protocol is less complex and needs less costly interfaces at Access Edge Node compared to Point- to-Point protocol approach (4) Operational savings are expected but not quantified issue within MUSE Phase II

Comparison: Cost break down – Ethernet (A-C) Migration

FTTCab (up to 50 Mbit/s Urban – about 30.000 customers connected)

10,0 Mio
5,0 Mio

0,0 Mio 5,0 Mio 10,0 Mio 15,0 Mio

2005 2006 2007 2008 2009 2010

Year Revenues Investments OAM Costs New Network functions implemented Cash Balance Default Case

New intelligent network functions enable significant CAPEX reduction

Investment in the Second Mile

0 Mio 1 Mio 2 Mio 3 Mio 4 Mio 5 Mio Basic Mixed PPP- DHCP BE=50% Multicast All Functions

Edge node Central Exchange Local Exchange

Implementation of multicast, QoS (e.g. 50% Best effort guaranteed), and mixed PPP/DHCP

Network Access Provider (NAP)

saves nearly 1/4 CAPEX within the aggregation network (second mile) by implementing new service enabling network functions

SLIDE 17

17

Major Results –

Migration Use Cases

Migration Case A-D – FTTC EPtP – NPV evolution comparison for various Areas:

Complete FTTCab: most profitable in dense urban, less profitable in urban

(1) Dense Urban, Urban, and (almost) Suburban are positive scenarios (2) Rural and Low Density Rural are negative scenarios NPV evolution

20 Mio
15 Mio
10 Mio
5 Mio

0 Mio 5 Mio 10 Mio 15 Mio

2005 2006 2007 2008 2009 2010

Reference A: Fibre to the Local Exchange (FTTLEx) up to 3/6 Mbit/s Dense Urban Urban Suburban Rural Low Density Rural

%&

%& %& %&

' ' ' '

%(

%( %( %(

) ) ) )

Reference today

%&

%& %& %&

%(

%( %( %(

SLIDE 18

18

Major Results –

Migration - Infrastructure Investment(CAPEX)

FTTLEx FTTLEx FTTC FTTH

Scenario

0,0 Mio 0,20 Mio 0,40 Mio 0,60 Mio 0,80 Mio 1,0 Mio 1,20 Mio 1,40 Mio Ethernet Aggregation 0,0 Mio 2,0 Mio 4,0 Mio 6,0 Mio 8,0 Mio 10,0 Mio 12,0 Mio 14,0 Mio Reference ATM Agg.

Comparison of Link Level Investments in Dense Urban areas

FTTLEx FTTLEx FTTC FTTH

Scenario

0,0 Mio 0,20 Mio 0,40 Mio 0,60 Mio 0,80 Mio 1,0 Mio 1,20 Mio 1,40 Mio Ethernet Aggregation 0,0 Mio 2,0 Mio 4,0 Mio 6,0 Mio 8,0 Mio 10,0 Mio 12,0 Mio 14,0 Mio Reference ATM Agg.

%. */ 0* %. %*/ 0*/ %. 10%*/ %. ' 01 %. 0' %. * 0

Link level investments of the access and aggregation network when comparing Fibre

to the Local Exchange (FTTLEx) to the Cabinet location (FTTC) is about 5 times higher.

For Fibre to the Home (FTTH) deployment is even in dense urban areas only an
ptimized roll out recommended

(1) The related link level investments

f the access and aggregation

comparing Fibre to the Local Exchange (FTTLEx) to the Cabinet location (FTTC) is about 5 times higher

(1)

(3) The FTTB solutions for Dense Urban needs about 30% less investments compared to the FTTH solutions (not shown). (2) In case of a complete FTTHome (GPON) deployment the link level related investments are nearly 10 times higher compared to the FTTCab case

(2)

SLIDE 19

19

FP1

In-house cabling

FP5

Local Exchange

FP7

Central Exchange

FP3

Cabinet Location

FP2

Building

FP0

CPE

FP4

Branch (1)

FP6

Branch (2)

First Mile Aggregation

IP- Backbone %* %* %* %*

RADIUS- Proxy DNS, Web- Cache

(FTTC)

DSL CPE

BRAS

GbE

Auto- configuration server

GbE

Switch Fibre Cable Twisted Copper Cable UTP5 Cable

Average link-lengths [m] LL1 (User ↔ Floor) 15 m LL2 (Floor ↔ Basement) 15 m LL3 (Basement ↔ Cabinet) 207 m LL4+LL5 (Cabinet ↔ Access Node) 1 060 m LL6 (Access-Node ↔ Aggr.-Node) 20 000 m LL7 (Aggr.-Node ↔ EN) 20 000 m

(FTTH EPON)

GbE

EPON ONU

Switch Optical Splitter (FTTH)

GbE

1G CPE

100- LX

Switch Switch

Use Case Definition –

Native Ethernet Use Cases – (1)

SLIDE 20

20

Major Results –

Native Ethernet Use Cases

Cash Balance Comparison first mile Native Ethernet “green field” scenarios

Complete FTTC: profitable in dense urban and urban area
FTTH: only optimised roll out even in dense urban recommended

Cash Balance (Urban)

60.0 Mio
50.0 Mio
40.0 Mio
30.0 Mio
20.0 Mio
10.0 Mio

0.0 Mio 10.0 Mio 20.0 Mio 30.0 Mio 2005 2006 2007 2008 2009 2010

60.0 Mio
50.0 Mio
40.0 Mio
30.0 Mio
20.0 Mio
10.0 Mio

0.0 Mio 10.0 Mio 20.0 Mio 30.0 Mio 2005 2006 2007 2008 2009 2010

Dense Urban

FTTC FTTBc (xDSL Basement Switch) FTTB (UTP inhouse ) FTTB ( fiber inhouse ) FTTH FTTH - EPON

SLIDE 21

21

Major Results –

Native Ethernet Use Cases –

High impact in the case of the Native Ethernet FTTH taking CPE costs into

account or not

Cost break down – FTTH incl. CPE

30.0 Mio
20.0 Mio
10.0 Mio

0.0 Mio 10.0 Mio 20.0 Mio

2005 2006 2007 2008 2009 2010

Year

Cost break down - FTTH (No CPE Costs)

30.0 Mio
20.0 Mio
10.0 Mio

0.0 Mio 10.0 Mio 20.0 Mio

2005 2006 2007 2008 2009 2010

Year Revenues Investments OAM Costs NPV Discounted Cash Balance

Native Ethernet (dense urban) - FTTH with and without CPE costs

SLIDE 22

22

Main Conclusions

Ethernet will substitute ATM in the Aggregation Area stepwise Functionalities moving closer to subscribers increased intelligence in access (service enablers like QoS, multicast, DHCP) Example A – C Migration (NAP): potential CAPEX savings of about 25% in

Aggregation

The techno-economic analysis supports the consolidated view on NG broadband access

Migration from a best effort access network (A) to a QoS enabled multi- service architecture based on Ethernet (A-C) or IP forwarding (A-D) are more economic than the ATM-based evolution scenario (A-B) Complete FTTCab: most profitable in dense urban, less profitable in urban FTTH: only optimized roll out even in dense urban recommended Native Ethernet “greenfield” scenarios show some advantages compared to the migration related scenarios In general the sensitivity analysis results support the key results that has been observed within the scenario comparison.

SLIDE 23

23