Más contenido relacionado La actualidad más candente (20) Similar a Market status report of IMS Issue 1 (20) Market status report of IMS Issue 11. ALAN QUAYLE BUSINESS AND SERVICE
DEVELOPMENT
M A RKE T S TATU S
R E PO RT O F IM S
( I P M ULTIM ED I A
S U BS Y S TE M )
AN INDEPENDENT AND QUANTIFIED VIEW
OF WHAT IS HAPPENING WITH IMS IN THE
TELECOMS INDUSTRY; INCLUDING A
GLOBAL SURVEY AND OPERATOR CASE
STUDIES.
© ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
2. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
CONTENTS
FORWARD 5
EXECUTIVE SUMMARY 6
OPERATOR RECOMMENDATIONS 12
NETWORK EQUIPMENT RECOMMENDATIONS 12
SERVICE LAYER COMPONENT (E.G. SERVICE BROKER) RECOMMENDATIONS 13
MOBILE HANDSET VENDORS 13
APPLICATION DEVELOPER RECOMMENDATIONS 13
INVESTOR RECOMMENDATIONS 13
INTRODUCTION AND BACKGROUND 14
PURPOSE 14
BRIEF IMS REVIEW 14
IMS MARKET SURVEY RESULTS 17
BACKGROUND ON THE RESEARCH 17
OPERATOR INTERVIEWEE ANALYSIS 17
SUPPLIER INTERVIEWEE ANALYSIS 19
IMS ACTIVITY 21
IMS BARRIERS 24
VENDOR PERCEPTION RANKINGS 27
IMS PLANS 28
IMS PRICING 30
RICH COMMUNICATIONS SUITE 31
ALTERNATIVES AND PATHS TO IMS 32
IMS MARKET SURVEY CONCLUSIONS 35
CASE STUDIES 37
VERIZON COMMUNICATIONS CASE STUDY 37
VERIZON BACKGROUND 37
PRINCIPLES BEHIND VERIZON’S IMS DEPLOYMENT 37
BREAKING DOWN THE SILOS 38
SERVICE DELIVERY ECOSYSTEM 40
INTEGRATION EXPERIENCE 41
CONCLUSION 42
CHINA MOBILE CASE STUDY 44
2 OF 66
3. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
VODAFONE SPAIN SERVICE BROKER 47
EXPLAINING THE SERVICE BROKER 47
VODAFONE SPAIN HOMEZONE SERVICE 48
APPTRIGGER CASE STUDY OF AN APAC OPERATOR 50
CASE STUDY REVIEW 53
WHERE THE MARKET IS GOING: IMS PROGNOSIS 54
CONCLUSIONS AND RECOMMENDATIONS 55
OPERATOR RECOMMENDATIONS 55
NETWORK EQUIPMENT RECOMMENDATIONS 56
SERVICE LAYER COMPONENT (E.G. SERVICE BROKER) RECOMMENDATIONS 56
MOBILE HANDSET VENDORS 56
APPLICATION DEVELOPER RECOMMENDATIONS 56
INVESTOR RECOMMENDATIONS 57
APPENDIX 1 – ACRONYMS 58
APPENDIX 2 – COMPANIES INTERVIEW 63
OPERATORS 63
SUPPLIERS 65
3 OF 66
4. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
TA BLE O F FI GU RE S
Figure 1. Operator IMS activity, and Distribution by Operator Type ____________________________ 7
Figure 2. Cumulative Percentage of Operators by year they anticipate starting IMS Deployment ______ 8
Figure 3. IMS Barrier and Frequency of Occurrence from Operators ___________________________ 8
Figure 4. Supplier Ranking of Barriers ____________________________________________________ 9
Figure 5. Estimate of the Initial Application of IMS _________________________________________ 9
Figure 6. Alternatives and Paths to IMS __________________________________________________ 10
Figure 7. IMS Architecture ____________________________________________________________ 15
Figure 8. Split between Operator and Supplier Interviews ___________________________________ 17
Figure 9. Geographical Distribution of Operators Interviewed ________________________________ 18
Figure 10. Operator Types Interviewed ___________________________________________________ 18
Figure 11. Distribution of Operator Interviews by Role ______________________________________ 19
Figure 12. Geographical Distribution of Suppliers Interviewed ________________________________ 20
Figure 13. Distribution of Supplier Interviews by Role ______________________________________ 20
Figure 14. Distribution of Supplier Interviews by Type of Supplier _____________________________ 21
Figure 15. Operator IMS Activity and Distribution by Operator Type ___________________________ 23
Figure 16. Live IMS Services __________________________________________________________ 24
Figure 17. IMS Barrier and Frequency of Occurrence from Operators _________________________ 26
Figure 18. Supplier Ranking of Barriers __________________________________________________ 26
Figure 19. IMS Vendor Ranking (Perception based)________________________________________ 27
Figure 20. IMS Vendor Ranking Regional Variation _______________________________________ 28
Figure 21. Cumulative Percentage of Operators by year they anticipate starting IMS Deployment ____ 29
Figure 22. Estimate of the Initial Application of IMS _______________________________________ 30
Figure 23. Potential Service Drivers _____________________________________________________ 30
Figure 24. IMS Price Range per Region (Cost per Subscriber in US Dollars) ____________________ 31
Figure 25. Rich Communication Suite Results _____________________________________________ 32
Figure 26. Alternatives and Paths to IMS _________________________________________________ 34
Figure 27. IMS Evolution _____________________________________________________________ 35
Figure 28. Verizon’s IMS and SDE Vision (source Verizon) __________________________________ 39
Figure 29. Shared Infrastructure Vision (source Verizon) ___________________________________ 41
Figure 30. Legacy Integration (source Verizon) ____________________________________________ 42
Figure 31. China Mobile Network Architecture (source Huawei) ______________________________ 45
Figure 32. China Mobile Target Architecture (source Huawei)________________________________ 45
Figure 33. Service Roadmap (source Huawei) _____________________________________________ 46
Figure 34. Typical Prior Situation (source OpenCloud) ______________________________________ 47
Figure 35. Service Broker Extending Pre-pay Applications (source OpenCloud) __________________ 48
Figure 36. Vodafone Spain Service Broker Scenario (source OpenCloud) _______________________ 49
Figure 37. APAC Operator Situation (source AppTrigger) ___________________________________ 50
Figure 38. Operator Decision Matrix (source AppTrigger) ___________________________________ 51
Figure 39. Proposed Solution (source App Trigger) _________________________________________ 51
Figure 40. Total Cost of Ownership Analysis (source AppTrigger) _____________________________ 52
Figure 41. Return on Investment Analysis (source AppTrigger) _______________________________ 52
4 OF 66
5. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
F O R WA R D
The religious war has ended, IMS (IP Multimedia Subsystem) did not become widely deployed
within the first decade of the 21st century. Surprisingly for many in the industry, SS7 (Signaling
System #7) remains in rude health and VoIP (Voice over IP) continues its linear, not exponential,
growth.
So where is IMS? Telenor has recently launched a suite of business services based on IMS.
Verizon and AT&T have been quite vocal on their IMS deployments. China Mobile and Telecom
Malaysia are both in the midst of IMS deployments. However, most of the field trials and initial
deployments did not migrate to full network deployments.
No one can argue against the need for IP multimedia session control with dynamic QoS (Quality
of Service), to enable 'reliable as the PSTN' voice communications across any capable network, and
for call features to work transparently across those disparate networks. So given the transition to IP
in most operators over the passed decade, why hasn't IMS followed that trend?
This report was stimulated from a series of frustrating conversations around IMS where there
appeared to be a significant gap between my experiences in working with operators around the world
in the services layer and the “reality” presented by a number of senior marketing people. This
document aims to provide a valuable independent status report that can help all in the industry have
a clear view on the current status of IMS and the likely paths operators will take in the evolution of
their networks.
5 OF 66
6. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
E X E C U T I V E S U M M A RY
During the months of July and August 2009, 137 phone interviews were conducted with
operators and suppliers around the world, 101 of those interviews were with operators. The
interviews consisted of a set of twenty open ended questions, designed to gather both hard and soft
data. That is, specific technical and commercial issues, as well as perceptions because both hard and
soft data influence decision making.
IMS remains niche, with only 8% of those operators surveyed deploying IMS, see Figure 1.
Note, none of those operators have completed the conversion of their network, all considered it a 5-
7 year process. Another 12% are in an extended field trial, which is characterized by services being
launched on the IMS core, with in some cases paying customers; but a decision has not yet been
made to commit to service migration onto the IMS core. Within this category are a group of mobile
operators that have deployed IMS for FMC voice, but do not as yet see a case for extending IMS to
other services.
IMS does not appear to be entering a period of rapid adoption, rather a linear growth in initial
adoption over the next 5 years, see Figure 2, with by 2014 about 32% of operators commencing an
IMS deployment. Regionally, NAR (North America Region) provides the bulk of the growth in years
2010 and 2011, while EMEA (Europe Middle East and Africa) and APAC (Asia Pacific) regions
provide the bulk of growth in later years.
Lack of „business case,‟ lack of „standards compliance‟ and „BOSS (Business and Operational
Support System) integration‟ were the top three barriers to adoption as identified by operators, see
Figure 3. The suppliers‟ views of the barriers were firstly a shorter list, see Figure 4; and different
rankings for the „Business Case‟ and internal to the operator „Network/IT organizational battle‟
barriers. Suppliers ranked „Business Case‟ fifth; instead rating technical issues higher. They ranked
the „Network / IT organizational battle‟ barrier higher than operators; citing their experiences in the
problems this posed in both gaining a decision from an operator for deployment, and then inflating
costs during implementation. Hence, significant IMS deployment challenges remain related to
internal operator organizational issues.
Over the past 3 years NEPs (Network Equipment suppliers) have struggled to gain market
interest in IMS, hence IMS prices have come down to the levels of softswitches at $2-3 per
subscriber. This is a significant change from the initial price points of $25 to $33 per sub quoted by
some vendors back in 2004/5 when IMS was first brought to market.
On the initial application of IMS, shown in Figure 5, the results show significant consensus
across operators and suppliers as well as across the regions. The top application is a „converged
VoIP platform across broadband, mobile and legacy.‟ Next was „end of life softswitch
replacement.‟ Operators, particularly in the US, are adopting this application of IMS as the price per
subscriber is trending towards $2.5 to $3 and their softswitches are reaching end of life. Also service
innovations such as HD Voice and mobile IP were not seen as drivers for IMS. Hence the value of
IMS value is seen in the core, not at the edge.
6 OF 66
7. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 6 shows the alternatives to IMS that were quoted by operators and suppliers interviewed.
Continuing with the existing IN and adopting a softswitch with its closed SCE (Service Creation
Environment) were considered top. Generally the qualifier around softswitch was its focused
application in the support of commodity voice. Which given the initial application of IMS is
anticipated to be by most operators as a core function in support of voice, rather than as a platform
for service innovation, demonstrates why most operators have not considered IMS until this year as
its pricing remained above that of softswitch.
Figure 1. Operator IMS activity, and Distribution by Operator Type
12%
Converged
25% Fixed only
Mobile only
63%
8%
Deployed, commercial 17%
services live
12%
Extended field trial,
limited commercial 44% Converged
service
Fixed only
Lab trial, under
evaluation 25% 58% Mobile only
Watching and waiting
36%
35%
45% Converged
Fixed only
Mobile only
20%
7 OF 66
8. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 2. Cumulative Percentage of Operators by year they anticipate starting IMS
Deployment
35%
32%
30%
26%
25%
22%
20%
17%
15%
12%
10%
8%
5%
0%
2009 2010 2011 2012 2013 2014
Figure 3. IMS Barrier and Frequency of Occurrence from Operators
Frequency of
Barriers occurrence
Business case 80%
Lack of standard compliance / proprietary
extentions 62%
BOSS integration 52%
Gaps on basic service emulation - lawful intercept,
ISDN2 48%
Destination without a route map 45%
Complexity 44%
General confusion 44%
Reaction to marketing hype 35%
Network / IT organizational battle 31%
Co-ordination between mobile, fixed and
broadband divisions 28%
Devices (mobile operator specific response) 25%
Economy 25%
Cheaper alternatives exist (Softswitch) 21%
Lacking web-based protocols 20%
8 OF 66
9. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 4. Supplier Ranking of Barriers
Frequency of
Barriers occurrence
Lack of standard compliance / proprietary
extentions 67%
BOSS integration 55%
Complexity 55%
Network / IT organizational battle 51%
Business Case 48%
Figure 5. Estimate of the Initial Application of IMS
Initial Application of IMS Frequency
Converged voice platform across broadband,
legacy and mobile networks 47%
Voice over LTE (Long Term Evolution) 8%
Voice over EVDO revA or HSPA+ 3%
Converged multimedia service platform across
broadband, legacy and mobile networks 2%
Service innovation, e.g. RCS (Rich
Communication Suite) 5%
End of life softswitch replacement 29%
Engineering bragging rights to maintain
stock/share price 6%
9 OF 66
10. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 6. Alternatives and Paths to IMS
Alternatives to IMS Frequency
JAIN SLEE 23%
SDP 42%
SIP Servlet and Service Broker 25%
Service Broker 28%
Softswitch 65%
Existing IN/SCP 76%
Paths being adopted Frequency
Softswitch for commodity voice no plan for IMS
adjunct in next 3 years 42%
Softswitch for commodity voice with IMS adjunct in
place or planned 29%
IMS core for VoIP services between mobile,
broadband and legacy (see Verizon case study) 18%
Continue to leverage existing IN assets (annually
reviewed) 47%
Service Broker (Vodafone Spain and AppTrigger
case studies) 23%
Focus on service exposure harnesses web
services SDP (SIP AS/Service Broker/ParlayX) 34%
Case studies on the IMS activities of Verizon, China Mobile, Vodafone Spain and an APAC
operator were reviewed. Verizon presents a leading deployment example of IMS adoption. Being a
converged operator provides a strong opex (Operational Expenditure) benefit in migrating to a
converged IMS core. Verizon‟s need to deploy LTE (Long Term Evolution, 4G) sooner rather than
later given the end of life of CDMA requires voice over IP support with QoS (Quality of Service)
over its RAN (Radio Access Network) before most other operators. Its high ARPU (Average
Revenue Per User) and a track record of technology leadership means it‟s able to tolerate the risk of
being an early adopter. Those characteristics are critical in driving Verizon to adopt IMS; we can see
a similar decision process for other non-GSM operators such as NTT.
For GSM operators the drive to LTE and IMS is not as strong, their 3G networks support
circuit based voice, HSPA+ (High Speed Packet Access) with the potential to deliver 42Mbit/s
within a 5MHz carrier delays their need to roll out 4G until 2013/2014, while CDMA operators are
limited to 3.1 Mbit/s. Some suppliers in the interviews made an argument that RAN efficiency will
drive all operators towards VoIP over LTE within the next 2/3 years. This argument is not reflected
in the operators concerns; their concern is solving the problem of cost effectively supporting mobile
broadband, not voice.
10 OF 66
11. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
In contrast to the conclusions drawn above, the China Mobile case study presents an interesting
scenario of a GSM mobile operator (or can be considered a relatively new converged operator)
making the move to IMS. Critical in its decision making is the need to offer business services over
its relatively new fixed broadband network. Hence it‟s in a relatively greenfield situation on the
broadband side so is rolling out IMS for its fixed line business, while also taking advantage of its
mobile service platform. The lack of a significant legacy fixed broadband network avoiding the issue
of scrapping sunk costs, and an attractively priced IMS solution appears to drive their decision to
adopt IMS.
Vodafone Spain provides an example of how a Service Broker is introduced to solve a problem
with current service delivery, which can then extend to manage the transition of services to SIP/IMS.
And finally the Apptrigger APAC operator scenario provides a quantified analysis of the financial
benefits in managing the transition from legacy to SIP/IMS through a service broker.
The case studies are just a few of the many diverse scenarios covered during this market study.
It‟s clear we‟re entering a phase of transition, where many more paths are being explored in the
transition to IMS, given the varied operator circumstance. Also in some cases aggressive IMS price
reductions have reached a point of parity with softswitch. However, the fundamental problem
remains of sunk costs in existing IN assets which will in most cases delay an operator‟s need to
evolve their network.
Summarizing the operator types and their propensity to adoption IMS based on the interviews,
from high to low:
Converged operator, CDMA based wireless infrastructure;
Converged operator, end of life softswitch or greenfield broadband with
„softswitch-equivalent‟ IMS pricing being offered from a NEP;
Fixed only operator; and converged operator where fixed and mobile divisions still
operate independently; and
GSM mobile only operator.
IMS is certainly not dead. The World remains „watching and waiting,‟ even given the
deployments in China Mobile, Verizon, AT&T and Telecom Malaysia to name just a few. From the
market survey we‟re seeing a slow linear adoption through the next decade, as shown in Figure 2.
Looking forward given the extended period of IMS adoption we could see several business
models emerge in its delivery. For example, hosted IMS; IMS is simply software that makes complex
session control decisions, hence a hosted model is feasible. For smaller operators given the higher
prices they are charged for IMS on a per subscriber basis; hosted IMS could enable better economics
by the time they consider deployment.
11 OF 66
12. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
OPERATOR RECOMMENDATIONS
GSM operators can afford to wait and see on IMS for at least 2 to 3 more years.
HD Voice can be supported on existing GERAN and UTRAN infrastructure
without IMS. HSPA will help mitigate mobile broadband economics, though it will
remain challenging.
Converged CDMA operators should follow Verizon, as CDMA technology will end
of life quicker than they expect, so they will need to manage the transition to LTE in
support of voice.
For fixed-line operators its really what deal can you cut with the supplier, if IMS is
roughly the same price as softswitch then it can make business sense if your
softswitch is EOL; as we‟re seeing with the cable companies in North America.
NETWORK EQUIPMENT RECOMMENDATIONS
Focus on converged CDMA operators, converged operators with an end of life
softswitch, and those operators where a clear Opex saving can be demonstrated
through the introduction of IMS into their core network. Avoid unclear service
innovation propositions to justify IMS.
Do not repeat the IMS marketing mistakes of the past in not acting in the
customer‟s best interest. At a critical time when operators need to be increasing the
rate of innovation, it was stalled through „brow beating‟ operators with a poorly
thought out and quantified IMS proposition.
Rein back standard development, its running too far ahead of the market. It can be
accused of “gilding the lily” rather than focusing upon market appropriate
standards. Focus on implementation guidelines, rather than yet more “bells and
whistles,” to aid deployment and integration as operators begin to deploy IMS.
Refresh the IMS marketing teams, they needs to be customer focused in helping
operators manage the transition, rather than brow beating operators with high-level
slideware.
Align business units to manage the transition to IMS, currently the business units
are acting as silos, making the transition for many operators difficult. Put simply,
focus on helping the customer, it will protect future shareholder value.
12 OF 66
13. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
SERVICE LAYER COMPONENT (E.G. SERVICE BROKER) RECOMMENDATIONS
Focus on both opportunistic legacy service problems, e.g. extending postpaid VAS
to prepay; and helping operators manage the transition to SIP/IMS through being
able to reuse their legacy service platforms and provide a degree of vendor
independence in the operator‟s choice of services.
Service layer components such as the Service Broker will be increasingly squeezed
by NEPs who have some / all of the functionality bundled in their IMS platforms.
Look to exit in next 2 years into a NEP or SDP provider.
MOBILE HANDSET VENDORS
A critical gap is IMS clients on devices. Today handset vendors look to operators to
pay. This creates a „chicken and egg‟ problem in extending IMS capabilities to the
edge. Operators, NEPs and handset vendors should work together to ensure the
IMS client (as well the HD voice codec) is just there, just like IPV6 is available on
most PCs even though its generally not used today.
APPLICATION DEVELOPER RECOMMENDATIONS
Only build IMS applications if the NEP or Operator pays as a consulting project, as
the application will likely be constrained to a particular platform.
Focus on the developer initiatives that provide direct customer access with open
web-based APIs.
INVESTOR RECOMMENDATIONS
IMS will be dominated by NEPs and will likely not see any hockey stick adoption,
rather a linear ramp through the next decade. Recommend focusing on service
opportunities that have both a direct to consumer / enterprise channel and through
an open telco channel. The service broker segment provides a near term
opportunistic investment.
13 OF 66
14. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
I N T RO D U C T I ON A N D BA C KG RO U N D
PURPOSE
This report provides an independent and quantified view of what is happening in the industry on
IMS (IP Multimedia Subsystem), through the presentation of results from an industry-wide survey
that encompasses 137 interviews, 101 of them being operators. The report also includes operator
and supplier case studies, presenting as factually as possible the current state of the art; without the
hype and marketing spin that has frustrated many people on this topic. These two objectives are
reflected in the two main sections of this report: market survey results and case studies.
An analogy that is frequently used during this survey is IMS presents a destination, however,
there's no map on how we get there from where we are today. This report aims to provide some
insight into the different paths operators are taking to aid operators in defining an appropriate
network strategy and to aid suppliers in making appropriate product and sales investment decisions.
This report is not an IMS primer, there are many good websites1 provide that function, and most
of the NEPs (Network Equipment Providers) will happily educate/indoctrinate operators on IMS.
Rather this report presents a level-set on IMS explaining and quantifying what people are thinking
and doing without the marketing spin that has clouded and confused the situation.
BRIEF IMS REVIEW
The IP Multimedia Subsystem (IMS) is an architecture for delivering Internet Protocol (IP)
based multimedia services with quality of service over multiple access networks from a common
core. It was initiated in mobile standards body 3rd Generation Partnership Project (3GPP), as a part
of the vision for evolving mobile networks. The original vision was the delivery of internet services
over GPRS (General Packet Radio System). That vision was subsequently updated by 3GPP, 3GPP2,
Cable Labs and TISPAN (Telecommunications and Internet Services and Protocols for Advanced
Networks) to support multiple access networks, such as Wireless LAN, CDMA2000 and fixed
broadband.
IMS uses IETF (Internet Engineering Task Force) protocols such as Session Initiation Protocol
(SIP), with some IMS specific extension, which has caused some frustration amongst the SIP purists.
IMS is not focused upon standardizing applications but rather defining a common horizontal control
layer for the access of multimedia and voice applications from wireless and wireline devices. Hence,
services do not require siloed control functions.
However, the architecture lacks implementation guidelines; in essence it is a destination without
a map on how to get there. Also it‟s a destination that is moving, with work items for release 10 of
the standard being discussed, while many mobile operators have only recently upgraded their core to
some components of release 5. It presents an unfortunate side effect of vendor dominated standards
bodies, they increasingly loose touch with the needs of operators. FSAN (Full Service Access
Network) was a good example of a body run by operators for operators, which created the
broadband passive optical network standard widely deployed today.
1http://www.rennes.enst-bretagne.fr/~gbertran/files/IMS_an_overview.pdf,
http://www.3gpp.org,
http://www.iec.org/newsletter/sept06_2/analyst_corner.pdf,
14 OF 66
15. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 7 shows a high level IMS architecture. Briefly explaining the main components:
SIP AS, Session Initiation Protocol Application Server. This runs all the logic for
the applications / services, e.g. IP Centrex.
S-CSCF, Serving Call Session Control Function. Call management and service
coordination. Provides the registration of the endpoints, routing of the SIP
signaling messages to the appropriate application server, and maintain session state.
The CSCF works with the interworking and transport layers to guarantee QoS
across all services.
P-CSCF, Proxy Call Session Control Function. Device control, interworking and
security. This is the entry point for IMS devices through an attached access
network, this routes the SIP messages to the user‟s home S–CSCF. It also manages
the quality of service requirements over the access network.
HSS, Home Subscriber Server. Stores all the user profile data.
BGCF, Breakout Gateway Control Function. Selects the network in which a PSTN
breakout is to occur, and interworking with the PSTN; its main role is IP level
peering for voice.
MGCF, Media Gateway Control Function. Manages the distribution of sessions
across multiple media gateways.
Media Server. Provides announcements, conferencing, etc.
Figure 7. IMS Architecture
IMS
SIP AS
HSS S-CSCF B/MGCF
P-CSCF Media PSTN
Server Interface
Multiple
PSTN
Access
Networks
15 OF 66
16. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Put simply IMS provides multimedia session control across multiple access networks with
standardized quality of service control. It enables an operator to have a common „core‟ across all its
networks for communication services (principally voice), and provides a relatively open environment
for value added communication services.
16 OF 66
17. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
I M S M A R K E T S U RV E Y R E S U LT S
BACKGROUND ON THE RESEARCH
During the months of July and August 2009, 137 phone interviews were conducted with
operators and suppliers around the world. The interviews consisted of a set of twenty questions,
designed to gather both hard and soft data. That is specific technical and commercial issues, as well
as perceptions because both hard and soft data influence decision making. The results of this survey
are presented in this chapter.
For the sake of transparency in this analysis, a notable interviewee exception is Ericsson which
would only provide public domain information. Ericsson‟s public relations department would not
allow any interviews. It is unclear why they need such control over information; this was in stark
contrast to the openness of the rest of the industry.
The split between operators and suppliers interviewed is show in Figure 8. The bulk of the
interviewees were operators. An objective of this report is to help guide operators in defining their
IMS strategy, hence gathering a broad set of data around operators‟ opinions and experiences enables
reasonable quantification of opinions at the level of the general market, region and operator-type.
The bias of the interviews also reflects the fact there are many more operators (750+) than suppliers
(5 global Network Equipment Providers (NEP)).
Figure 8. Split between Operator and Supplier Interviews
Suppliers
26%
Operators
74%
OPERATOR INTERVIEWEE ANALYSIS
The geographic distribution in Operators interviewed is show in Figure 9. Given the diversity of
operator situations, the interview process attempted to provide as broad and representative a
geographic coverage as possible. A notable feature in the operator interviews was many operators
thought other regions where more advanced in the deployment of IMS than their own. A finding
that will be explored in more detail later is North America has the broadest deployments and that
situation is likely to remain for several years due to specific market factors that will be discussed later.
17 OF 66
18. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 9. Geographical Distribution of Operators Interviewed
Americas
24%
APAC
38%
EMEA
38%
The split between operator types interviewed is shown in Figure 10, the definitions used are:
Converged: within a country of operation having both fixed and mobile operations;
Fixed only: for example broadband ISPs (Internet Service Providers)
Mobile only: within the country of operation.
If an operator is planning to launch a mobile offer to its existing fixed line business they are
considered a converged operator for this analysis because it impacts their decision making and
business case around IMS. For example, a cable operator that is planning a mobile operation is
considered a converged operator in this analysis.
Figure 10. Operator Types Interviewed
Mobile only
39%
Converged
49%
Fixed only
12%
18 OF 66
19. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 11 shows the distribution of Operators interviewed by role. The aim was to gather a
representative mix of roles across the organization. This was done to uncover any differences in
opinions based upon role within an operator or uncover intra-organizational factors. There is a slight
bias towards the technology office, in part because they are the group responsible for assessing IMS
so in some cases only the technology office was prepared to offer time for an interview.
Figure 11. Distribution of Operator Interviews by Role
IT
20% Technology
Office
33%
Network
Engineering
22%
Marketing
25%
SUPPLIER INTERVIEWEE ANALYSIS
Figure 12 shows the geographic distribution of the suppliers interviewed. Global refers to
supplier interviewees whose scope is across all markets, e.g. global marketing. While some of those
interviewed also are focused upon a specific geography. This was captured to determine any regional
differences in suppliers‟ opinions. APAC refers to Asia Pacific; and EMEA to Europe, Middle East
and Africa.
19 OF 66
20. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 12. Geographical Distribution of Suppliers Interviewed
APAC
22%
Global
38%
EMEA
22%
Americas
18%
Figure 13, shows the distribution of interviewee roles. Again the objective was to achieve a
representative scope across the supplier organization to see if there were any organizational
differences. For example, marketing‟s role is to generate demand, so they will have a view slightly
different to that of the technology office or account management.
Figure 13. Distribution of Supplier Interviews by Role
Sales /
Account
Management
23%
Technology
Office
44%
Marketing
33%
20 OF 66
21. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 14 shows the distribution of supplier interviewees by company type: NEP (Network
Equipment Provider, e.g. Ericsson or Huawei), network component supplier (e.g. service broker or
SBC (Session Border Controller), e.g. AppTrigger), BOSS (Business and operations Support Systems,
e.g. Amdocs), device/edge component software supplier (e.g. mobile device or femtocell gateway).
Figure 14. Distribution of Supplier Interviews by Type of Supplier
Device/Edge
component
19%
NEP
43%
BOSS
19%
Network
component
19%
IMS ACTIVITY
This section presents the general findings from the market research. Figure 15 shows the status
of operator IMS activities. Note the percentages shown cannot be extrapolated to the general
market, because by the nature of the interview process, i.e. operators being willing to give their time,
the bias is to those operators active or considering IMS. Though with that said, 95% of the
operators approached were willing to be interviewed; only 5% responded that they either had no IMS
plans so would not interview or were not prepared to discuss their IMS plans.
IMS is deployed in 8 operators interviewed, out of the 101 interviewed, with the majority of
those being converged operators. The main driver for a converged operator is the opex gains as the
IMS core can be shared across the fixed, mobile and broadband access networks in support of VoIP
services. The criteria for an operator to be in this category is live commercial services are being
migrated onto the platform.
21 OF 66
22. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
The second category, extended field trial, of which there are 12 operators in this category that is
differentiated by the fact service migration has not yet started. Specific services have been launched
on the IMS core, with in some cases paying customers; but a decision has not yet been made to
commit to service migration. As a result the IMS platform essentially remains a silo; for example
several mobile operators have used IMS with their FMC (Fixed Mobile Convergence) voice service,
but do not see a business case for migrating other services onto IMS. In this category converged
operators still dominate. In about half the operators anticipated moving towards deployment, but
the timing was uncertain.
The third category of lab trial refers to an IMS platform being evaluated within a lab, generally by
the technology office of an operator, 36% of operators interviewed were in this category. It is
interesting to note that roughly half the operators in this category received their IMS platform for
free as either an evaluation system, or as part of another deal; for example several operators had
recently made a softswitch order, and an IMS platform was bundled in that deal. Roughly 80% of
the operators in this category were unsure about when they would move towards deployment.
Finally, the largest category is “watching and waiting,” accounting for 44% of operators
interviewed. That is, there is no IMS platform under active evaluation; though there may have been
such an evaluation in the past. The operator does not see the business case for deploying IMS, and
hence is continuing business as usual. Once the business case becomes clear they would make the
move. The majority of mobile operators are in this category.
Within this „watching and waiting‟ category is a group of operators who questioned whether IMS
in its current form would be relevant when they finally consider deployment; for example several
brought up ideas such as hosted IMS. We‟ll consider this in a later chapter.
Given the limited data set for deployments it‟s difficult to draw any geographic trends, but North
America would appear to be leading in deployments and planned deployments. Some of the market
factors considered to be the main driver for this situation are:
Converged operators are in the process of consolidating their broadband, mobile
and legacy core networks / operations;
End of life softswitch infrastructure, and NEPs offering core IMS platforms at
comparable prices;
Converged operators dependence on end-of-life CDMA which leads to the early
adoption of LTE (Long Term Evolution) and the need to support VoIP support
over DORA (EVDO (EVolution Data Only) RevA) to understand the implications
of mobile VoIP, to enable LTE to adequately support VoIP; and
Relatively uncompetitive market with higher than average margins which enables
strategic technology decisions to be taken, that in more competitive markets would
be difficult to justify.
22 OF 66
23. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 15. Operator IMS Activity and Distribution by Operator Type
12%
Converged
25% Fixed only
Mobile only
63%
8%
Deployed, commercial 17%
services live
12%
Extended field trial,
limited commercial 44% Converged
service
Fixed only
Lab trial, under
evaluation 25% 58% Mobile only
Watching and waiting
36%
35%
45% Converged
Fixed only
Mobile only
20%
Of those operators that have deployed IMS or are in an extended field trial the services supported are
shown in Figure 16. Voice emulation (this includes FMC voice) is by far the most popular, i.e. using
the platform to support voice (PSTN) services over IP, followed by IP Centrex, which is generally
targeting the business market. PTT (Push To Talk) was more common in the extended field trial
group, than the deployed group.
23 OF 66
24. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 16. Live IMS Services
100%
80%
60%
40%
20%
0%
SIP Trunking
Push to Talk
Voice emulation
Communications
IP Centrex /
Unified
IMS BARRIERS
Given the maturity of IMS, and the move most operators have made to IP in their core
networks; IMS remains niche, as is clear from Figure 15. Figure 17 shows the list of barriers quoted
by operators on why they have not deployed IMS. Lack of a clear business case has been leveled at
IMS for some time. A more accurate view may be a lack of a common business case, as operators
must migrate from their existing situation, which means there is no one simple business case.
Take for example the business case for NGN (Next Generation Network); it can be easily
justified by the cost savings from simply closing down the multiple SDH (Synchronous Digital
Hierarchy) silos built over the years. Of those operators who have started to deploy IMS a common
factor is being a converged operator with the need to support voice services over multiple access
networks. Their specific situation made IMS a much easier choice than say a GSM mobile operator.
Lack of standards compliance and/or vendor proprietary extensions were raised by over half the
operators interviewed, though to be fair to the NEPs many of the non-compliances are due to
differences in the interpretation of the complex IMS standards. Most went as far as to say that they
would likely adopt a solution from one vendor given the challenges and costs incurred in managing
integration. This calls into question the scope of the IMS standard. A point raised by many of the
suppliers of products that integrate to an IMS core is operator require IMS compliance on the RFQs,
but then do not enforce compliance in the deployment, raising significant unexpected integration
costs. This is compounded by the fact that because IMS is a software solution it tends to sits
between network engineering and IT organizations within an operator; it‟s essentially a software
integration project managed by network engineering, which tends to result in many expensive change
orders.
24 OF 66
25. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Surprisingly many operators in their experimentation with IMS discovered it could not
adequately support legal intercept or ISDN2 emulation. This is a significant barrier for operators to
start migrating services. Legacy service support is always a thorny issue, as how far should the IMS
platform go in supporting services that perhaps would be better made end-of-life. Given IMS has
only recently achieved a level of maturity that has allowed early adopter operators to start migrating
services, this gap is to be expected, but will like constrain the scope of the more aggressive IMS
deployments.
Even the most anti-IMS operator still regards IMS as an inevitable destination, because no
adequate standards-based alternative exists to provide IP based multimedia session control with
quality of service. However, a common complaint was the lack of a map from today, their currently
network situation, on how to migrate to an IMS core. Later we will review some of the paths
operators are taking, see Figure 26, and map them to operator situations to see if there are some
common characteristics.
Rather than the IMS spec becoming simplified and more focused over time with deployment
experience; more and more capabilities have been added, Release 10 is currently being defined; this
leads to the complexity barrier. In part with a lack of broad based deployment experience, there was
little feedback into the standards body, which should have perhaps taken a holiday for a few years to
let the market catch up. Some operators consider the separation of the I-, S- and P-CSCFs an
architectural mistake; where a simplified approach would have saved cost and implementation
complexity.
IMS has been presented in a number of incarnations over the years, initially presented as the
method to support instant messaging and push to talk when those services were popular amongst
operators. Once the PTT trials demonstrated customers weren‟t that interested in the services at the
price necessary to cover the cost of the investment, the focus moved to service exposure and
operational savings; which were difficult at best to prove. The focus then moved onto the fixed
operators, linking it closely to the NGN transformation, generally with a service innovation spin.
The NGN cases proved out, however, the IMS case did not. And of the past year the focus has
changed yet again to IMS supporting IPTV, which is discussed in this weblog article2. This shifting
proposition has resulted in general confusion.
Reaction to the hype surrounding IMS cannot be understated. This factor caused by far the
most venting from operators, even more than vendor proprietary extensions to IMS standards.
Operators felt badgered by NEPs, being asked to believe not think. The blame for this is laid
squarely at the NEPs marketing departments. This has resulted in a backlash. A lesson to the
industry is B2C marketing doesn‟t work, it‟s a B2B business, hence work with your customers and
don‟t treat them like „punters‟ to be „duped‟ into the „latest fashion.‟
The prioritization of barriers was common across regions, operator types, and role within the
organization. Lack of business case, lack of standards compliance and BOSS integration were the
top three barriers across the regions.
2 http://www.alanquayle.com/blog/2009/06/why-is-ims-getting-lumped-into.html
25 OF 66
26. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 17. IMS Barrier and Frequency of Occurrence from Operators
Frequency of
Barriers occurrence
Business case 80%
Lack of standard compliance / proprietary
extentions 62%
BOSS integration 52%
Gaps on basic service emulation - lawful intercept,
ISDN2 48%
Destination without a route map 45%
Complexity 44%
General confusion 44%
Reaction to marketing hype 35%
Network / IT organizational battle 31%
Co-ordination between mobile, fixed and
broadband divisions 28%
Devices (mobile operator specific response) 25%
Economy 25%
Cheaper alternatives exist (Softswitch) 21%
Lacking web-based protocols 20%
When Suppliers ranked the barriers, shown in Figure 18 the main differences were firstly a shorter
list with much less emotion attached to those barriers; and different rankings for the „Business Case‟
and „Network/IT organizational battle‟ barriers. Suppliers ranked „Business Case‟ fifth; instead
focusing upon technical issues. However, they did rank the „Network / IT organizational battle‟
barrier higher than operators. Citing their experiences in the problems this posed in both gaining a
decision from an operator for deployment, and then inflating costs during implementation.
As described earlier a broad base of suppliers were interviewed, not just the NEPs. This ranking
reflects the problems they are facing in integrate with IMS. In particular BOSS vendors reported the
greatest number of problems even on what should be relatively mature diameter interfaces across
both provisioning and billing. The difference between the supplier and operator barrier rankings
reflects their positions in the value chain. However, there is a significant gap around the business
case which suppliers will need to fill to move operators to action on IMS.
Figure 18. Supplier Ranking of Barriers
Frequency of
Barriers occurrence
Lack of standard compliance / proprietary
extentions 67%
BOSS integration 55%
Complexity 55%
Network / IT organizational battle 51%
Business Case 48%
26 OF 66
27. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
VENDOR PERCEPTION RANKINGS
Now back to operator views, operators were asked to rank the main IMS vendors on a scale of
1-5 with 3 being industry average; based upon their perception of the NEP‟s IMS solutions. Note
this question was not trying to provide a detailed feature by feature comparison of which NEP‟s IMS
solution is best; rather provide a higher level insight into how operators ranked the vendors. In
Figure 19 the aggregate totals are shown.
I‟ve also broken the rankings down into regionally in Figure 20. There is the usual bias of NEPs
performing better in the „home‟ regions. However, Huawei came top across all regions, even though
Huawei has not been selected as often as say Ericsson. It is important to note that the smaller
operators in a region carry the same voting power as a large national operator in this survey. So even
though a traditional NEP may be selected by the large incumbent operator; the smaller operators,
which appear to be a target for Huawei, helped Huawei to the top of the rankings.
Figure 19. IMS Vendor Ranking (Perception based)
4.00
3.44
3.50 3.15
3.12 3.02
2.78 2.88
3.00
2.50
2.00
1.50
1.00
0.50
0.00
ALU Ericsson Huawei Motorola NSN ZTE
27 OF 66
28. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 20. IMS Vendor Ranking Regional Variation
4
3.5
3
2.5 APAC
2 Americas
1.5 EMEA
1
0.5
0
ALU Ericsson Huawei Motorola NSN ZTE
APAC 2.98 3.14 3.64 2.65 3.21 3.14
Americas 3.21 2.93 3.29 2.89 2.95 2.55
EMEA 3.16 3.38 3.39 2.79 2.91 2.95
IMS PLANS
Figure 21 shows when operators anticipate starting an IMS deployment. There does not appear
to be a knee, rather a steady growth over the next 5 years. Regionally, NAR provides the bulk of the
growth in years 2010 and 2011, while EMEA and APAC provide the bulk of growth in later years.
As discussed previously, the EOL (End Of Life) of CDMA, focus on LTE, and most operators
having converged networks means IMS has a stronger proposition for NAR operators. EMEA and
APAC mobile-only operators remain laggards with >50% anticipating deployment beyond 2015.
Later we will review some of the initial IMS deployments in the case studies chapter, for example
Verizon, China mobile, and Telecom Malaysia.
28 OF 66
29. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 21. Cumulative Percentage of Operators by year they anticipate starting IMS
Deployment
35%
32%
30%
26%
25%
22%
20%
17%
15%
12%
10%
8%
5%
0%
2009 2010 2011 2012 2013 2014
Operators and suppliers were asked what they through the initial application of IMS would be,
shown in Figure 22. The results showed significant consensus, across operators and suppliers, as
well as across the regions and other dimensions. The top initial application was considered to be a
“converged VoIP platform across broadband, mobile and legacy.” Next was end of life softswitch
replacement. Several operators particularly in the US are making this move as the price per
subscriber for IMS is trending towards $2.5 to $3 (everything but the HSS). So IMS‟s principle value
is seen in the core, not at the edge.
Beyond the top two applications, where IMS is not taken to the edge of the network, its clear
IMS is not being considered for service innovation. For mobile operators device issues dominate.
Even for broadband operators, their experiences with IMS clients and SIP stack interoperability
issues have made them rather shy of taking IMS to the edge. For service innovation the SDP would
appear to have occupied that spot, even though ParlayX is part of IMS. IMS is seen to lack the
necessary web-based protocols and integration, hence why operators are looking towards IT vendors
such as Oracle for delivery of SDP solutions.
29 OF 66
30. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 22. Estimate of the Initial Application of IMS
Initial Application of IMS Frequency
Converged voice platform across broadband,
legacy and mobile networks 47%
Voice over LTE (Long Term Evolution) 8%
Voice over EVDO revA or HSPA+ 3%
Converged multimedia service platform across
broadband, legacy and mobile networks 2%
Service innovation, e.g. RCS (Rich
Communication Suite) 5%
End of life softswitch replacement 29%
Engineering bragging rights to maintain
stock/share price 6%
Extending the previous question into what specific services could potentially be a driver for IMS,
see Figure 23, this backs up the results shown in Figure 22 on the perceived lack of service
innovation as a driver for IMS. Postpay VAS on prepaid had strongest interest in APAC (Asia
Pacific region) and LATAM (Latin American region), though in most cases the operator was not
considering full IMS, rather an interim step towards IMS in the use of a service broker as discussed
in the Vodafone Spain and AppTrigger case studies.
Figure 23. Potential Service Drivers
Service Driver Frequency
HD Voice 0%
LTE 6%
Mobile VoIP 6%
RCS 8%
Open Innovation 10%
Postpay value added services on prepaid 15%
Operator App Store 8%
Other 5%
IMS PRICING
Figure 24 shows the IMS pricing ranges quoted by operators. Note these prices do not include
HSS (Home Subscriber Server), but do cover every other function and the integration necessary to
have the IMS core in an operational state, e.g. for some operators they quoted it as their „turnkey,‟ or
„build, operate, transfer‟ price. This is a significant change from the initial price points of $25 to $33
per sub quoted by some vendors back in 2004/5 when IMS was first evaluated.
Over the past 3 years NEPs have struggled to gain market interest in IMS, hence prices have
come down to the levels of softswitch, at $2-3 per subscriber. In the APAC operator interviews 22%
of them claim to be evaluating IMS on either a platform provided for free, or bundled in a softswitch
sale. Hence, it is shown as zero as it‟s claimed the license agreement would allow commercial traffic
to flow.
30 OF 66
31. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
There is always some difficulty in comparing quoted prices as some deals have IMS positioned as
an adjunct to a softswitch, rather than standalone. The biggest determinant on price was the size of
the operator, the smaller the operator the higher the price.
Figure 24. IMS Price Range per Region (Cost per Subscriber in US Dollars)
Region Low High
APAC 0 3.9
EMEA 2.5 5.4
NAR 2.5 4.8
RICH COMMUNICATIONS SUITE
RCS3 (Rich Communications Suite) is a GSMA initiative with the purpose of speeding up and
facilitating the introduction of commercial IP Multimedia Subsystem (IMS) based rich
communication services over mobile networks for both UMTS and CDMA initially and later
extending to fixed networks. The focus of RCS is upon the user experience, interconnection and
interoperability requirements tied to a core feature set of rich communication capabilities. The
GSMA are positioning RCS as an important driver for the introduction of IMS.
RCS‟s features include:
Enriched Call:
Interactive communications. E.g., content (video/Image) sharing, file transfer
Multimedia ID presentation; info of the caller is pushed to Callee (from device
or network based)
Enhanced Messaging
Conversational messaging
Single entry point for all messaging services
Communications log
Mobile and Fixed environments
Enhanced Phonebook
Service Capability Indication (video call, One-shot / Ad-Hoc group messaging,
File transfer)
Start voice call, video call, File Transfer, Messaging from phonebook
3 http://www.gsmworld.com/our-work/mobile_lifestyle/rcs/gsma_rcs_project.htm
31 OF 66
32. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Customize Multimedia ID presentation and save Multimedia ID details on a call
establishment
Presence info (Publish, authorize other users to see, display other‟s presence)
Allow searches locally and externally (yellow/white pages, corporate directories)
Back-up/Synch with network address book
Figure 25, shows the results of questioning around RCS across both operators and suppliers.
Only one third of the interviewees were actively tracking IMS, surprisingly 21% were not aware of
RCS, they were generally from marketing departments which should be the target customers for RCS
within operators, and technology office had the best awareness.
Nearly half the operators interviewed had awareness but did not consider it significant enough to
track. On deeper questioning around operator‟s view of RCS, the majority of operators considered
RCS had a lack of web integration, citing web alternatives as already available and with a slicker user
experience, this as also reflected in the suppliers views. In roughly one third of the interviews
operators were examining integration with web based alternatives, rather than RCS, e.g. Yahoo! IM.
Half of the operators interviewed considering it an attempt to maintain legacy business models and a
political attempt to limit Skype‟s attempts to lobby for open access.
Figure 25. Rich Communication Suite Results
Aware of RCS?
Tracking RCS 35%
Aware not tracking 44%
Not aware 21%
View on RCS Frequency
The break-through service IMS needed 18%
A user-centric, integrated experience, that finally
moves the address book into the 21st century 24%
A political ploy to limit Skype's attempts at open
access 43%
Lacking web integration 75%
An attempt to maintain legacy business models 55%
ALTERNATIVES AND PATHS TO IMS
So if operators are not adopting IMS, then what are they doing? Figure 26 shows the alternatives
that were quoted by operators and suppliers interviewed. Continuing with the existing IN and
adopting a softswitch with its closed SCE (Service Creation Environment) are considered top; which
in general means business as usual. Generally the qualifier around softswitch was its focused
application in the support of commodity voice. Which given the initial application of IMS is
anticipated to be by most operators as a core function in support of voice, rather than as a platform
for service innovation, see Figure 22, demonstrates why most have not started to make the move
until this year as IMS pricing remained above that of softswitch.
32 OF 66
33. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
SDP was considered a partial alternative, generally for service innovation. The term SDP4
requires further definition, the Verizon case study provide a good example of the roles of IMS and
SDE (Service Delivery Ecosystem, which falls into the category of SDP). Beyond that JAIN SLEE,
SIP Servlet (with Service Broker) and Service Broker were considered alternatives to IMS, though in
practice they are stepping stones which allow existing capitalized assets to continue generating highly
profitable revenue in support of value added services (not switching infrastructure). In the case
studies we‟ll delve into this transition in more detail.
After asking about alternatives, the next question focused on the paths operators are taking
towards IMS, note these paths are not mutually exclusive, so in some cases operators are following
multiple paths. For example, some operators were adopting an SDP for service exposure, while also
adopting softswitch for commodity voice services. Slightly under half the operators interviewed were
unsure on which path they would take, so the percentages do not apply across all operators, only of
those who expressed an opinion. I‟ve also taken a certain amount of interpretation in the grouping
of the paths, as the specific implementations are quite diverse.
Leverage existing IN assets accounted for nearly half responses, which had a close correlation to
those operators who were watching and waiting. Next was using softswitch (NGN) for commodity
voice with no plan for IMS in the next 3 years, these options are not mutually exclusive as roughly
half those interviewed selected both. Skipping over service exposure, as that option is relatively
independent of the others; deploying an IMS adjunct to a softswitch was in third place. Essentially
using IMS for more complex services, generally business services, followed closely by use of a service
broker to reuse existing service silos on SIP/IMS.
4http://www.alanquayle.com/blog/2008/05/defining-soa-sdp-and-ims-and-h.html
http://www.alanquayle.com/blog/2008/08/service-delivery-platform-land.html
33 OF 66
34. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 26. Alternatives and Paths to IMS
Alternatives to IMS Frequency
JAIN SLEE 23%
SDP 42%
SIP Servlet and Service Broker 25%
Service Broker 28%
Softswitch 65%
Existing IN/SCP 76%
Paths being adopted Frequency
Softswitch for commodity voice no plan for IMS
adjunct in next 3 years 42%
Softswitch for commodity voice with IMS adjunct in
place or planned 29%
IMS core for VoIP services between mobile,
broadband and legacy (see Verizon case study) 18%
Continue to leverage existing IN assets (annually
reviewed) 47%
Service Broker (Vodafone Spain and AppTrigger
case studies) 23%
Focus on service exposure harnesses web
services SDP (SIP AS/Service Broker/ParlayX) 34%
Figure 27 tries to map out these alternatives into a roadmap. The service broker is broken out as
a separate element, even though most IMS adjunct and IMS core implementations include broker
functionality, to highlight the case were an operator has decided on a third party for the service
broker to provide some vendor independence on where their services come from. Those highlighted
in red were most popular.
The current transition is taking the industry into a more complex situation, were essentially an
operator can add an IMS adjunct to a softswitch to support more complex services, or can transition
out the softswitch to an IMS core. But regardless that legacy SCP keeps hanging on in there, though
there are a few exceptions.
Generally those operators implementing a service broker bring it in to solve problems around
making postpaid VAS available to prepay customers. Or legacy service silos are still profitable so the
operator wants make those services available across its IP based voice services. Or to solve a specific
service interaction problem as described in the Vodafone Spain case study.
The options shown in Figure 27 overly simplifies the situation as some operators have not yet
integrated their fixed and mobile operations, so this adds another dimension to the evolution. And
the evolution also ignored the BOSS integration, the approach taken by Verizon (see later case study)
is the introduction of a mediation layer, but this approach is not yet being followed by the rest of the
industry, instead the approach is custom integration which adds significant costs tin managing the
transition.
34 OF 66
35. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 27. IMS Evolution
IMS Adjunct
Service
Softswitch
Broker
Legacy
SCP
Legacy
SCP
IMS Adjunct
Softswitch Softswitch
Service
Legacy Broker Legacy
IMS Core
SCP SCP
Current
Softswitch Transition IMS Core Service
Legacy Legacy Broker
SCP SCP
IMS Core
Legacy
SCP
Summarizing the operator types and their propensity to adoption IMS based on the interviews,
from high to low:
Converged operator, CDMA based wireless infrastructure;
Converged operator, end of life softswitch with softswitch-equivalent IMS pricing
being offered from a NEP;
Fixed only operator; and converged operator where fixed and mobile divisions still
operate independently; and
GSM mobile only operator.
Geography and the country‟s economic conditions do not appear to play a role in the operators‟
propensity to adopt IMS. Rather whether they benefit from consolidating onto a common IMS core
to achieve operational savings dominate.
IMS MARKET SURVEY CONCLUSIONS
IMS remains niche, with only 8% of those operators surveyed deploying IMS, see Figure 15.
With another 12% in an extended field trial, this is differentiated by the fact service migration has not
yet started, specific services have been launched on the IMS core, with in some cases paying
customers; but a decision has not yet been made to commit to service migration. IMS does not
appear to be entering a period of rapid adoption, rather a linear growth in adoption over the next 5
years, see Figure 21, reaching 32% of operators by 2014 starting an IMS deployment. Regionally,
35 OF 66
36. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
NAR provides the bulk of the growth in years 2010 and 2011, while EMEA and APAC provide the
bulk of growth in later years.
Lack of business case, lack of standards compliance and BOSS integration were the top three
barriers to adoption as identified by operators. When suppliers ranked the barriers, shown in Figure
18, the main differences were firstly a shorter list with much less emotion attached to those barriers;
and different rankings for the „Business Case‟ and „Network/IT organizational battle‟ barriers.
Suppliers ranked business case fifth; instead focusing upon technical issues. However, they did rank
the „Network / IT organizational battle‟ barrier higher than operators. Citing their experiences in the
problems this posed in both gaining a decision from an operator for deployment, and then inflating
costs during implementation.
Over the past 3 years NEPs have struggled to gain market interest in IMS, hence prices have
come down to the levels of softswitch, at $2-3 per subscriber (excluding HSS). This is a significant
change from the initial price points of $25 to $33 per sub quoted by some vendors back in 2004/5
when IMS was first evaluated.
On the initial application of IMS, shown in Figure 22, the results showed significant consensus,
across operators and suppliers, as well as across the regions and other dimensions. The top initial
application was considered to be a “converged VoIP platform across broadband, mobile and legacy.”
Next was end of life softswitch replacement. Several operators particularly in the US are making this
move as the price per subscriber for IMS is trending towards $2.5 to $3. So IMS‟s principle value is
seen in the core, not at the edge.
Figure 26 shows the alternatives that were quoted by operators and suppliers interviewed.
Continuing with the existing IN and adopting a softswitch with its closed SCE (Service Creation
Environment) are considered top. Generally the qualifier around softswitch was its focused
application in the support of commodity voice. Which given the initial application of IMS is
anticipated to be by most operators as a core function in support of voice, rather than as a platform
for service innovation, see Figure 22, demonstrates why most have not made the jump until this year
as IMS pricing remained above that of softswitch.
Figure 26 shows the alternatives that were quoted by operators and suppliers interviewed.
Continuing with the existing IN and adopting a softswitch with its closed SCE (Service Creation
Environment) are considered top; and even when an operator starts to implement IMS, its generally
introduced incrementally, with converged operators leading the way.
36 OF 66
37. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
CASE STUDIES
VERIZON COMMUNICATIONS CASE STUDY
VERIZON BACKGROUND
Verizon is an incumbent US broadband and telecommunications company, a converged
operator. It was formed in 2000 when Bell Atlantic, one of the Regional Bell Operating Companies
(RBOCs), merged with GTE. Consolidation within the US market has resulted in two quad-play
(voice, video, broadband, and mobile) telcos: Verizon and AT&T. Generally they operate in
different regions for residential broadband and telephony, so only directly complete on
wireless/mobile services and in business services. Within their incumbent fixed regions the main
competition comes from cable operators, who deliver a triple play proposition of voice, video and
data; and are now expanding into wireless services.
Verizon‟s latest results published in July 2009 shows 2.5M FiOS (video) customers with an
average ARPU of $135 per month (blend of voice and/or video and/or data). While retail ARPU is
at $72.00. Across their 87.7 million wireless customers, 80 million or 91% of their base is retail
postpaid, with a low churn of 1.37% per month. Data revenue growth was an annualized 33.2% with
non-messaging services up 44% and messaging up 20%; and data ARPU increased to $15.00.
BlackBerry users at Verizon have an average ARPU of over $100. Verizon‟s EBITDA was $6.2
billion for the last quarter, up 9.4% from last year on a pro forma basis.
Compared to many operators Verizon has the benefit of high ARPU and low churn. And in the
fixed space a duopoly with the cable provider. Satellite does provide a degree of competition for
video services, though with customers‟ increased awareness and use of interactive video services such
as VoD (Video on Demand) and catch-up TV, satellite is increasingly uncompetitive.
In this case study I will review the role of IMS and Verizon‟s SDE (Service Delivery Ecosystem)
in delivering their N-Screen services vision. This is done to clearly demonstrate the role of IMS; and
the additional technology requirements to achieve service innovation.
PRINCIPLES BEHIND VERIZON‟S IMS DEPLOYMENT
Ubiquitous Broadband
Broadband networks are the foundation of Verizon‟s strategy; they are investing to meet their
customers‟ needs for speed and mobility, for delivering their digital experiences anytime, anywhere,
and to whatever screen they have at hand. Specifics include evolving to 4G/LTE for wireless
devices, and continuing to build out FiOS/FTTP, and their global IP network.
37 OF 66
38. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Convergence: Shared IMS
Convergence is about technology consolidation; bringing their networks, devices and processes
under one centralized control/management to improve efficiency and enable a unified customer
experience including ordering, subscribing, billing, management, and execution of services (which are
no longer tethered to a particular network or device.) IP Multimedia Subsystem (IMS) is their chosen
technical solution for enabling applications to work across their multiple networks, and across their
customers‟ multiple screens.
Integration: Services Delivery Ecosystem
Integration is about applications and services inter-working, cooperating and leveraging the
user‟s digital experience, which is across both equipment and subscribed applications and services.
Integration is how Verizon delivers the benefits of converged infrastructure to their customers.
Verizon‟s Services Delivery Ecosystem (VZSDE) is their technical solution for enabling the rapid
development and delivery of integrated applications.
Innovation: Open Development Initiative
To improve the benefits of convergence, Verizon are working to tap into the innovative energy
of their customers, suppliers, and partners; and to leverage the combined power of IMS, the Web,
and attached devices/networks. This involves building the infrastructure as well as new developer-
focused business processes to enable rapid service development and delivery on our converged
networks. They have defined an Application-to-Network Interface (ANI) to expose key enablers
(location, presence, conference, profile, address book, etc.) such that they are easily incorporated by
developers into innovative new services; examples include Verizon‟s Open Development Initiative
(ODI) and Verizon‟s Developer Center5.
BREAKING DOWN THE SILOS
Figure 28 shows Verizon‟s vision in how IMS and SDE break down the existing silos to enable a
shared converged infrastructure across its fixed, broadband and wireless access networks. The main
business driver for Verizon in this migration is operational expense reduction.
5 http://developer.verizon.com/blogs/
38 OF 66
39. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 28. Verizon’s IMS and SDE Vision (source Verizon)
Voice Voice Voice Msg Msg Msg
PSTN Wireless IP PSTN Wireless IP
Communications Entertainment Information
IMS + SDE
Global IP Network
FiOS LTE
Verizon‟s vision is services will be available across all its networks. Verizon‟s N-Screen vision
results in customers considering their services as independent of a particular network: whether it be
mobile, broadband or legacy networks. Services can roam between those networks, or be transferred
as required by the customer, with a common security framework; finally removing multiple logons
and conflicting security settings which plague most multiplatform services today.
Media within sessions can be added / dropped as required; e.g. a customer viewing an interest
clip on Verizon broadband can share that with a friend on Verizon FiOS to view on their TV, or
with another friend on their smart-phone over the mobile network. From a customer‟s perspective
it‟s just drag and drop, with the network managing the complexity.
IMS and SDE enable common service components such as security, charging, and media
adaptation to be re-used across services; removing the need for multiple gateways. Coupling this
with the Development Ecosystem enables Verizon to „mash-up‟ their services with the web. As well
as opening up the network to non-traditional wireless devices, this goes beyond M2M (Machine to
Machine), rather enabling many more consumer electronic devices to securely interconnect over the
Verizon network.
39 OF 66
40. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
SERVICE DELIVERY ECOSYSTEM
The principles guiding the development of Verizon‟s service delivery ecosystem are:
Create Innovative Services: The ecosystem needs to support a wide variety of
developers in order that creativity as well as utility is afforded an opportunity to
reach customers.
Increase service velocity: Applications and services must be easy to build and able to
use and reuse common building blocks. Applications and services must also be
readily and easily integrated into the business and operations support systems.
Share infrastructure across multiple lines of business: Sharing and reuse of
infrastructure within the ecosystem occurs at several levels: from the actual
development environment tools and facilities, to the network resources (i.e.,
building blocks) that support the applications and ultimately to the management
components that make the applications available to customers.
Grades of service from “Telco quality” to best effort: Not all applications and
services, nor do all customers demand the highest quality for every scenario.
Applications and services need to be tunable to the needs and willingness to pay of
the customer.
One of the primary drivers for the Verizon Service Delivery Ecosystem (VZSDE) is the ability
to enable rapid development and delivery of new services and service suites to consumer and
business customers.
Verizon aims to offer applications that highlight their unique strengths and capabilities such as
multimedia communications, wireline and wireless network connectivity and video content
distribution. Verizon is able to offer Quality of Service (QoS) for services delivered on the Verizon
network which is likely more difficult for “over-the-top” service providers. Since Verizon can reach
the same customer in multiple ways, Verizon wants to tap into their unique ability to share common
customer data and provide context and continuity across the various access devices and methods to
enhance the user‟s experience with Verizon.
The service creation capabilities within the VZSDE support a diverse development community.
This development community ranges from trusted, in-house developers, to contracted third party
developers and ultimately independent third party developers. All should be able to build applications
that use Verizon network and/or device capabilities.
The VZSDE will support Verizon‟s vision of an innovation lab framework. This framework
provides a flexible process for application developers to use to create and test new service concepts
and for Verizon to use to evaluate, assess and promote promising new applications and services.
40 OF 66
41. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
INTEGRATION EXPERIENCE
Part of the business case justification in deploying IMS was the creation of a single/shared view
of the customer. Enabling the support of services independent of device or network, creating a
stronger quadruple play proposition, for example customers can consume video services via their
mobile phone, PC, or STB. This is a critical point in the rationale for Verizon‟s IMS deployment,
operational expenditure savings drive the initial deployment, and then new service revenues will drive
its scale.
Figure 29 demonstrates Verizon‟s view that enablers are broader than just IMS (e.g. ParlayX),
and should be available to all applications and services; as well as application partners. Hence the
important role the SDE plays in aggregating enablers within a common policy framework. The
enablers also extend outside the development community to enterprises, so they can mash-up
presence from its enterprise networks with presence of its employees on the Verizon network.
The ecosystem is also able to augment the enablers provided by VZ‟s SDE, allowing developers
to mash-up additional services, to create innovative web-centric enablers that make it easier for
applications to use network capabilities without being a telecoms expert.
Figure 29. Shared Infrastructure Vision (source Verizon)
IMS is not replacing Verizon‟s network one day; rather it is an incremental process very much
linked to initially operational savings and then new service revenues. An “anchor tenant” creates the
case for its introduction; in Verizon‟s case LTE and FiOS. And over time services are migrated onto
that platform when economics make sense, e.g. when an existing service platform reaches its end of
life. Hence the need for BOSS mediation so the traditional market-facing business units can
transparently manage this migration, as shown in Figure 30.
41 OF 66
42. © ALAN QUAYLE BUSINESS AND SERVICE DEVELOPMENT
Figure 30. Legacy Integration (source Verizon)
For Verizon, 3GPP Release 8 closed significant gaps in the IMS specifications around
convergence and multimedia support that had led it to define A-IMS (Advanced IMS,) their version
of the IMS. The benefits of R8 include IP mobility to support VoIP services over the wireless
network, particularly handover between cells. Note the mobile operator Three‟s Skype phone still
uses circuit based voice over the RAN. Note Verizon is a CDMA operator, so their 3G networks
lack the circuit based voice support found in GSM networks. This is a key driver for their early
adoption of IMS, which we will review in more detail later.
Additional benefits accrued from R8 include an evolved packet core, common IMS, and service
continuity; all necessary for the support of voice over packet in both EVDO and LTE radio access
networks. However, challenges still remain in areas such as converged policy management,
transcoding, IMS video, emergency services, session shifting, common subscriber database and media
plane security.
CONCLUSION
IMS builds on existing Web capabilities with new enablers like contextual awareness, network
control, and a single notion of user. IMS is just a component, in the limit it provides multimedia
session control with real-time quality of service; essential for supporting voice over packet across its
multiple networks.
Verizon‟s Developer Initiatives, aimed at fostering innovation requires a SDE. Verizon has one
of the clearest views on how IMS and SDP work together; integrating communication networks and
the Web. Creating simple things like being able to associate a web transaction with a voice call, a
small innovation with the potential to significantly impact commerce, security and improved user
experience.
A second component of the Developer initiatives is to accelerate innovation, and integration
(mashing up services). Enabling rapid service delivery (concept to commercial) to bring Verizon
more inline with the innovation rates of web-based services providers.
42 OF 66