Broadband Multi-Service Networks: The advance to IP networks of the future
Broadband Multi-Service Networks
This paper describes how the business development of network operators is driving the evolution of conventional communications networks to broadband multi-service networks. The main conclusions of the paper are that: • The fundamental development driver is a person’s need and desire to communicate with other people through any available communications medium. This means that any multi-service solution must be able to offer real-time, conversational communication services. Furthermore the need for high-speed access to information on a global basis is an additional major driver. • The advent of new technology, and major changes in market structure, stimulate the development towards multi-service products and networks. Successful business and network development is based on utilization of the legacies in competence, infrastructure and customer relationships. This is especially valid for large operations. • • • An evolutionary network development is the most pragmatic strategy. The primary criteria for investment in a multi-service infrastructure is financial: technology is a secondary criteria. Two primary business models, telephony centric and the datacom centric, are the starting points of a migration strategy towards multi-service networks. Both models often reside with the same network operator. • • • A flexible, multi-service network structure is horizontally layered. Connectivity, communications control and applications are kept separate. A multitude of communication and other services will be offered, deployed and managed through the multi-service network. For network operators, Ericsson’s suite of ENGINE solutions offers a successful roadmap to the multi-service marketplace.
Page 2 of 21
Broadband Multi-Service Networks
Broadband multi-service networks – a definition
Ericsson defines a broadband multi-service network by its three key characteristics: it is capable of handling communication services for voice, data and real-time conversational multimedia; it is capable of handling bandwidth-demanding applications and content; it embraces access and aggregation, switching/routing, communications control and platforms for online applications and content.
An industry in turmoil
In the last decade, legislation and new regulations have imposed sweeping changes on the telecommunications industry, transforming it from a supply-driven to a demand-driven environment. Consequently, enterprises and consumers have learned to behave in new ways that immediately impact upon the services requested from, and offered by, service providers. Although damaging to profits and stock prices, the current downturn in the telecommunications industry should not be seen as evidence that the compelling forces and fundamental changes that we describe here have been overrated or misunderstood. We firmly believe in a new telecoms world that will realize increased productivity and enhanced quality of life for its end users. The present difficulties in the telecommunications industry derive from a combination of two factors: a global economic downturn; the symptoms of an over-hyped new economy.
The latter is characterized by speculation in ever-increasing stock values and, among other factors, is based on business assumptions about bandwidth requirements, technological choices and network presence (the larger the network the better). However, the desire of an individual to interact with others cannot be overlooked, or underestimated. It is a fundamental and ever present need – unaffected by fluctuation in any industry.
Communication needs increase and change
The Internet has become a major force on the economy, on society, and on individuals. The most enthusiastic pundits of the “new economy” claim that the Internet will change everything and, although this assertion can now be disputed in detail, it is clear that the Internet is a powerful facilitator of change. The impact of the Internet – including Mobile Internet – has accelerated the ordinary evolutionary changes in the way that society behaves. The societal and behavioural changes most likely to affect the telecommunications industry can be categorized as those that take place in business enterprises and those that occur among individual end-users. Enterprise behavior • Globalization. Today, a growing number of enterprises operate in a global environment, with suppliers and/or customers in other countries. In many cases the company itself has become a global entity: directly, or with partners in an international cluster.
Page 3 of 21
Broadband Multi-Service Networks
Restructuring, outsourcing and clustering. Increased competition compels companies to focus on their core businesses. It is also imperative to keep capital costs low by employing “just in time” logistics strategies. Thus, many more inter-company connection points are created, increasing the need for personal and system communications that are rapid, efficient, reliable and information heavy.
Towards knowledge based organizations. As the educational standard of employees increases and decision-making becomes more distributed and delegated, internal communications within enterprises assume even greater strategic importance. Companies are putting more effort into keeping their employees up to date – with on-line conferences, web-casting, internal TV-channels, newsletters, multi-location workshops and many other means of disseminating information. To address the above considerations, enterprises need effective communication solutions that
support individual productivity as well as the interaction with customers and business partners. It is obvious that improved, widespread, person-to-person communication capabilities are a key factor for successful development. Further increases in individual productivity require local and wide area mobility, with secure, any-time/anywhere access to information and applications. Thus employees need: • • • Communication and computing devices, e.g. a mobile phone, a laptop and maybe PDA, equipped to communicate and synchronize, and possibly also a fixed phone at the desk. Service & application packages personalized for the duties and job of the individual. Remote access when on the move or working from home.
The behavior of end-users • People want and need to interact with others. In a telecommunications context, this fundamental behavior is served by telephony (voice and video), e-mail, web chat rooms and similar means. • Consumers are embracing the “experience society”. Shopping malls, movie theatres, commuter networks, etc. have become meeting places with different attractions for all tastes. In experiments, some meeting places are being developed into “full service community centers” that cater for private and public services alike. They must be connected within a global communications network that provides consumers with the attractions that they seek, at the site and from home. • With higher standards of living, consumers now demand “convenience”, “personalization” and “freedom”. In telecom terms, these demands translate into mobility related services such as the ability to get in touch at any time, and always have access to information on the move, exemplified by the multitude of new, home-based services now being tested. Initially, many users will take up broadband for its immediate benefits of always-on, fixed price, high speed Internet access. It is also less inconvenient than the relatively longer call set-up and waiting times inherent in dial-up narrowband, particularly when phone lines are occupied. Subsequently, consumers will be attracted by “next generation” services that include: broadband enabled entertainment services such as on-line games, personalized broadcast radio and video (“Personal TV ”); applications hosting and conversational multimedia; always-on support for the intelligent home of the future, and its many communicating appliances and applications (e.g. security).
Page 4 of 21
Broadband Multi-Service Networks
All these growing demands have certain common factors. They require rapid access to other people and information, at high quality, and with the feeling of “being there”. Reduced quality or significant delays have an immediate negative impact for the subscribing enterprise and/or consumer, who do not then perceive the service as efficient or convenient to their needs.
The competitive environment in change
Few industries have undergone such rapid upheaval as that recently experienced in telecommunications. The hitherto stable, regulated market has been dramatically altered by deregulation, re-regulation, technological development, new services and globalisation. Traditional service providers once controlled the entire value chain and were able to achieve high margins. Then they encountered the new conditions of a de-regulated market – new services, new technology, and huge investment capital available to potential competitors. Established business concepts were scrutinized and new, horizontal models were created in parallel with the conventional, vertically-integrated architecture. Fresh competitors such as telecom brokers, access providers and pure wholesale companies emerged. The market was divided, and a service provider could focus on just one segment – like international voice or dial-up Internet. Services such as mobile telephony and Internet subscription attracted many users while, at the same time, the high margins in international and long-distance calls drew many new entrants into the market. Risk capital was easily available. The market value of a network operator was based on expectations of a future customer base and network capacity. This led to major investment in networks and the customer base – initially from newcomers in niche areas, then by incumbents throughout the market. At first the investment was compatible with growing demand. However, in a second phase, a “bubble” was created when networks were built on speculation of massive future demand, especially in the area of long-haul transmission. Growth in the customer base was achieved by price competition, so the once profitable long-distance traffic soon became a low-margin business. New entrants were hit especially hard.
Today’s network is characterized by low tariffing and over-deployment in long-haul transmissions. Many newcomers have failed; their assets acquired by larger players. The market is consolidating. Margins for voice services are decreasing due to competition. In the search for new sources of income, players from other industries are fighting for a slice of the cake. Cable TV operators, content providers and consumer-centric industries, with or without strong brands, see the opportunity to bundle their conventional products with a different kind of communication service. New value systems are being established. Cost-effectiveness has become an even more significant competitive factor than usual. At the same time there is a need for new, value-adding services that can create new revenues, but at limited risk. Any business strategy adopted by a network operator will address these issues in this order. The operator must have a cost efficient network that offers high quality voice, data and multimedia services.
New network requirements – new technologies
It is generally accepted that the next generation of networks will be IP-centric and multi-service. In comparison with other technologies, IP scores highly on a number of strong features. IP as an applications enabler – applications drive network use and development. End-to-end IP transport will enable users to access any IP application. The very openness of the architecture makes it easy to introduce and prototype new applications. IP as driver of cost reduction – the decoupling of applications and control from the underlying infrastructure allows for the optimization of different types of products with different drivers. Costs will be reduced because IP is the focus of mainstream product and technology investment and development. However, IP in itself is basically a connectionless, best-effort technology that was not designed for voice or any other real-time service. To best exploit the benefits stated above, a new generation of IP must be developed and implemented in the networks.
Complexity, insecurity, risk
The most striking characteristic of the telecoms industry today is that network operators and service providers are facing uncertainties that affect complex decisions that have far-reaching consequences. There is market-wide confusion as to where, when, and how to make the next move. As discussed above, a number of unpredictable forces have impacted upon operators. They include fundamental changes in the behaviour of residential and business subscribers, increased competition resulting from deregulation/consolidation, and the impact of upcoming new technologies. But what really keeps network operator CEOs awake at night is the fact that margins for existing, legacy-based services are decreasing while the revenue-generating capabilities of services based on new technologies remain unknown. Network operators and vendors who want to create pro-active, viable strategies must keep this dilemma at the forefront of their minds. For further analysis, we have grouped these uncertainties into three categories: legacy, finance and technology.
Page 6 of 21
Broadband Multi-Service Networks
Legacy consists of the operator’s existing customer base, competence and infrastructure and can be seen as an asset and a liability. At the peak of the last economic boom much of the industry, vendors and operators alike, mainly perceived their legacy as a liability. However those with a more realistic view continued to develop a migratory approach, using existing customer relationships, competencies and networks while moving towards the new telecom world. Customer relations Existing customer relationships offer the best opportunities for network operators to succeed, but they also represent the hardest challenges. Traditional operators have access to most of the consumers and business in the market. They have well-established relationships that can, and should, be developed. But there is a significant downside. These same customers do not expect their telecom and datacom providers to offer more than just the basics. Business subscribers will not seek new, value-adding services from their old, familiar network operator. Instead they will turn to an advertising agency for help with the development of customer relations; they’ll go to a technical consultant to improve their supplier connections; they’ll take on a management consultant to solve internal communications problems, and so on. Similarly, in their search for new “experiences” and greater “convenience”, private consumers will go to media companies, grocery stores and other such providers. The danger in this scenario is that the established network operator will spiral down the value chain, providing only high volume/low margin bandwidth. Other players will own the all-important, value-adding, customer relationships. A network service provider with ambitions to grow beyond a basic carrier service needs to address these issues by building, acquiring or partnering with strong brands in the respective areas. Competence The competence legacy derives from when the industry was technology driven and based around an existing network paradigm. The key factor then was the core technical competencies of the network operator – such us the ability to develop, design, build and run networks with Quality of Service, realtime capabilities and charging/accounting functionality. In a multi-service environment however, mere technical competence is not enough. New technologies and new business models must be introduced into existing organisations. And the quality demands of customers do not decrease when new technology is introduced, rather the opposite. Endusers require better service quality at a lower price, thereby imposing upon the network operator the need for constantly increased competence. Infrastructure The infrastructure legacy differs, depending on the origins of the service provider. A telephony-centric organization has invested in a circuit switched network whereas a datacom-centric organization has invested in IP, Frame Relay or a similar network. In all cases there is a need to utilize the existing infrastructure while transforming or migrating the business from a single service to a multi-service entity. The reuse of investments is essential for economic reasons as well as the need to keep the existing service active. But ultimately it is too expensive to maintain the existing services in a legacy network without migrating to a multi-service network. The existing infrastructure would not exist if it had not been effective in addressing its target applications and customer requirements. Any generational shift in technology must be able to handle a
Page 7 of 21
Broadband Multi-Service Networks
particular application at least as well, and preferably better, than the existing solution. Alternatively, the need for the service delivered by the existing technology must be made redundant by a new business model. Both of these arguments have been advanced: for instance in the IP telephony v PSTN debate. But with today’s generation of IP technology, delivery of a full set of PSTN services is actually more expensive than existing circuit-switched or call control solutions that utilize ATM connectivity. There is also the argument that voice services should be free, as put forward by some new players, but this cannot be valid until new revenue-generating services and their associated business models actually exist. As long as telephony remains the main revenue generator, the income derived from it cannot be put at risk. Instead, as margins continue to fall, operators must focus on delivering existing services at lower cost. For the service provider with a legacy network, this can only mean one thing – a cost efficient migration strategy to a broadband multi-service network.
As always, finance is a major consideration. Risk capital is available only for the most solid business cases. Competition has depressed prices and threatens the profitability of existing services. Operational costs and wages are all increasing, and growth in traffic exceeds that of revenue by a number of magnitudes. Subscriber demands for customized services are also driving costs upward. Together, these factors compel the successful communications service provider to adopt a step-bystep approach when building a cost-efficient, broadband, multi-service network. To succeed, the service provider must cater for new demand, re-use existing investment, secure new revenues and savings, and make new investments at the same pace.
Operators contemplating broadband multi-service networks are confronted by a number of technological issues: the technology must support the optimization of network resources; it must support the easy, flexible and cost efficient deployment of new services; it must be a roadmap to the all-IP network of the future.
However, when making choices about technology, the most important stance is to ignore the hype and begin to see the technology as a means – not an end in itself. IP will certainly be the dominant strategic protocol of the future but, in current implementations, especially for certain mission-critical services, there are obvious limitations at present. Choice of technology must be made – based on a sound business case, not hype or future expectations.
Our findings so far …
To summarize, operators face a number of uncertainties: • They are experiencing a paradigm shift in business and technology, leading many to continuously question the nature and validity of their core business. In turn this gives rise to perpetual reorganization and internal change. • They are uncertain as to how changes in the behavior of end users will translate into demands for new service, and which customers will lead the take-up of new services.
Page 8 of 21
Broadband Multi-Service Networks
Many are undecided as to which technologies will finally become the industry standard for different parts of the network and thereby offer significant cost benefits, and in what time-frame. The favorable capital markets have stimulated the establishment of new operations and growth, intensifying the competitive situation to the point where it has become difficult to describe, or even monitor. From all these circumstances, a number of network-related considerations emerge: quality of real-time services in terms of latency, delay and bandwidth; openness of different services and brands to sharing the infrastructure; possibility for the growth of competence with that of service; possibility for continued use of investments that have already been made; continuance of existing services without loss of quality or feature content; possibility to quickly reach new customers with new services; possibility to make big leaps in terms of cost cutting. openness that makes it possible to adapt a strategy as existing uncertainties are resolved and new ones occur.
A pragmatic advance to the all-IP multi-service network
To meet the many challenges that were have discussed here, operators must approach them from many angles. To become more market oriented, flexible and cost efficient are obvious objectives: so is the development of new business models supported by a service-independent network infrastructure. Another fundamental finding is that a new architecture, a “next generation network”, is needed, but it is also evident that this must evolve. A leapfrog strategy that ignores legacy services, customers and infrastructure is not viable. The necessary investment in new services must be funded by the delivery of existing services.
A layered architecture In a traditional network architecture, different voice and data services are integrated with the particular technology that transports and switches them. The network is separate, and vertically integrated. Service convergence, transport and switching integration, and the implementation of new services are achieved only by great effort. To address the heterogeneous requirements, a new architecture is being agreed upon within the telecommunications industry. The result is an open structure in which communications functions are separated into the following, broadly defined, horizontal layers: • • • • Content and user applications that reside at the edge of the network, accessible through the network services. Network resident communication control applications, broadly comprising: an upper layer that controls service provisioning and usage, e.g. service access, mobility, AAA functions, etc., a lower layer that controls network infrastructure resources.
Connectivity, i.e. packet transport across backbone and access networks.
Separation of control and connectivity is the essential characteristic of the new architecture.
Page 9 of 21
Broadband Multi-Service Networks
The main benefits of the layered architecture are: • • • • It supports an evolutionary approach to network development, allowing the migration of technologies as the need arises, thus reducing risk. It is technology agnostic because the fundamental architecture is independent of the technology chosen for the different layers It is service independent, providing connectivity and control for legacy and new services, thus enabling the rapid deployment of new services It enables superior traffic engineering capabilities.
Packet Backbone Network Mobile Access (3G)
Copper Access Fiber Access
Fixed Radio Access
A layered architecture offers an evolutionary approach to network development, facilitating the rapid migration from one technology to another at need. It thus becomes an efficient tool for overcoming the myriad uncertainties about customers, services, traffic demands and technologies discussed at length above. Quality of service capabilities There can be little doubt that current concerns about the security and service quality of IP networks will be solved eventually with international effort. The most important and time-consuming aspects of IETF standardisation work are the increased number of parties involved, and considerations of mobility. Market consolidation and a reduction in the number of players may be beneficial in this regard. But although IP QoS and security considerations will be solved by standardization, the solutions may prove to be complex and difficult to handle efficiently. To be effective, the standards must be implemented in all involved network elements, in a communication flow from source to destination through a number of interconnected networks.
Page 10 of 21
Broadband Multi-Service Networks
QoS considerations may be managed by implementing “IP traffic engineering”, which imposes a connection-oriented mechanism on to best-effort technology. This incurs complexity costs that will be addressed with the upgrade of the IP network. Over provisioning of network resources is another way to manage QoS, but control mechanisms and some traffic engineering will still be necessary. An efficient approach that offers a pragmatic, evolutionary way forward is one in which: network solutions are based on the layered network architecture; they are open to the inherent benefits of an all IP solution; and technologies are chosen according to their current merits.
Whatever the solution chosen for managing real time traffic in an IP environment, the importance of revenue streams from traditional voice and corporate data communication services cannot be overemphasized. In our view, any network solution must protect these sources of revenue, with particular emphasis on quality of service, network availability and security.
Legacy network and business models It is important to adopt a pragmatic approach to migration and network evolution. A new network cannot be constructed from scratch, at least not profitably. To achieve goals of customer, revenue and competence retention, the chosen strategy must be that of network evolution or migration. Different technologies should be employed when designing network solutions. They depend on circumstances and customer needs, but must always focus on the business objectives of reduced costs, increased revenues and positive cash flow. It is also necessary to strive for flexible, open solutions that facilitate changes of technology on any layer of the network architecture, including a general evolutionary route towards an ultimate packet-based, IP-centric network. It is especially important to keep in mind that, when migrating from a vertical network structure, it is not only the different operators who have diverse requirements. In fact different types of operation exist even within a single operator – characterized by different business models, service offerings, customer base and infrastructure. Today there are two types of vertical, network operations that will migrate to a multi-service network. One originates from the telephony business and is referred to here as a telephony centric operation. The other originates from private and public datacom services and is here called a datacom centric operation. For operators who are entirely telephony centric or datacom centric, the business case can be analysed purely from that perspective, and network evolution can be designed and optimized relatively easily. But the majority of network operators run both types of operation, so a thorough study of network requirements is essential. It must address methods of building a multi-service network that take telephony centric and datacom centric requirements into account, otherwise the full business case will not materialize. Telephony centric operations In telephony centric operations: a substantial portion of the business and the most of the revenues come from telephony; the subscribers are used to 99.999 % quality of service as well as an extensive range of services (PSTN, ISDN and IN);
Page 11 of 21
Broadband Multi-Service Networks
the existing infrastructure that provides the services is predominantly TDM/SDH, with close relationships to narrowband access and legacy OSS; the customer base is residential and commercial; the service is often regulated, with minimum delay requirements and constraints such as Lawful Interception.
The dilemma for managers of this type of operation is that prices for the existing offerings are being continuously eroded, while no new sources of revenue are visible in the near future. The main business objective here is to lower the cost base while still delivering existing services transparently in terms of reliability, quality and feature content. These goals include a restructuring of the access network from narrowband to broadband and, eventually, the delivery of new types of services over the network. Datacom centric operations In datacom centric operations, the existing business is centered on legacy datacom services – mainly managed FR and LAN interconnect, bandwidth, leased lines, Internet access and similar, low-content applications; the customer base is predominantly commercial, but usage by residential subscribers is slowly growing through broadband access solutions; the existing infrastructure is based on packet technologies such as IP and Frame Relay, often running over an ATM core. The main problems for datacom centric operations are the handling of network traffic growth, price pressure on existing services, and the creation of more revenue from the present infrastructure. Consequently the business objectives for network evolution are low cost network expansion and the capability to increase the number of services and applications. From the start, datacom centric services included voice (but probably not the full-feature telephony of the legacy telephony network), and the relatively simple multimedia services that can be supported by H.323 based server solutions. However, streaming services such as multicasting are now in demand. Also, more sophisticated multimedia services will be added to this segment as technology evolves, in the connectivity layer and in the actual control and service creation mechanisms. A combined strategy Although the requirements of telephony centric and datacom centric operations differ to some extent, from the overall perspective of the operator it is critical that they can be fulfilled within the same network solution. This must utilize the same backbone and access infrastructure and enable service inter-working between the network domains and between subscribers. Cost savings are essential in telephony centric operations. As a consequence, the initial migration strategy is based on ATM, which guarantees the quality of the services and applications. Later the strategy moves to IP when multi-protocol label switching (MPLS) and differentiated services (DiffServ) can guarantee service quality and traffic engineering for a large customer base. The telephony services are controlled by centrally located telephony servers, which not only provide transparent services to end users but are also transparent in terms of the relationships with access and legacy OSS. It should also be noted that, during the initial ATM based phase, MPLS can also be used for fast, efficient IP transport over the ATM network, before IP is finally introduced as the connectivity technology.
Page 12 of 21
Broadband Multi-Service Networks
In the next phase, multimedia services can be added by introducing a SIP based multimedia control server. Further utilization of the legacy infrastructure can be achieved by upgrading the existing copper access networks with DSL technologies and direct connections to the media gateways. It is important to avoid a technology dead-end by ensuring that the ATM equipment used in the first phase of the strategy has a smooth migration path to IP. To transport traditional voice telephony over IP networks, an ATM switch can be transformed into a multi-service switch/router. This is done by introducing IP routing boards into the ATM chassis, and by re-using existing processors, switch cores and backplanes. In the next phase ATM will move to the edge, being replaced by IP in the core – leading to an all-IP network when the telephony access has been moved to broadband access. For datacom centric operations, the expansion of the service to existing customers includes the introduction of voice services in the office environment, with the option to progress to multimedia services such as network-based, high quality, video conferencing. The business case for voice services is cost based because it is predicated upon utilization of the existing office infrastructure and appliances. A full set of telephony services is not required (as distinct from the telephony centric discussion above), but the quality aspects remain the same. The path of the datacom centric solution is built around IP telephony servers that provide enhanced voice services over the IP network. Also in this scenario, real-time multimedia services will be added by the inclusion of SIP-based multimedia control servers in the architecture. Metropolitan access solutions are a key element in this path. The migration to IP is not only driven by service expansion, but also by increased focus on cost due to the price pressures on existing services. The customer expansion path is towards small office, home office and consumers through Internet access. The first step is modem pool subscriptions. The second step has own broadband access deployment connected to the metropolitan networks and driven by the business from enterprise users. The telecom centric and datacom centric paths must inter-work because their legacies exist and the services must go between them. But in the next step the paths converge. We propose that those operators who have the different requirements described above should take a dual approach to the multi-service network. They can differentiate their service offerings and business models for different market segments, but still realize considerable cost savings from a common infrastructure. This solution will not merely safeguard income until new, revenue-generating services roll out. Studies have shown that operators who take this route gain a time advantage over those who wait for IP technology to mature enough to support a direct step from TDM to IP. In the end it is not a matter of technological preference as much as having a positive bottom line.
The architecture of the new multi-service network must support a multitude of services. They range from legacy telephony and datacom services to new, innovative multimedia applications, and from best-effort to real-time. Conventional telephony Conventional telephony will remain a key service for some time to come, because of its large user base revenues. Service providers who have a large customer base in conventional telephony will hesitate to risk the integrity of their service. Nevertheless, over time, users will increasingly migrate to mobile and IP alternatives for their voice communications.
Page 13 of 21
Broadband Multi-Service Networks
Conventional telephony has been the core business of large operators for more than 100 years. Its circuit switches are ubiquitously deployed throughout the wireline and mobile networks. TDM switching technology is one of the most embedded in the network – with its integrated application, control and connectivity planes and intimate relationship with access and legacy OSS. To implement large scale TDM switch replacement, operators must consider the risk to revenues from the inability to deliver income-generating local and transit telephony services. The substantial costs of access and OSS uncoupling must also be taken into account. Recent standardization specifies a development of the traditional circuit switch, taking the functionality into the packet world and utilizing new signalling protocols such as H.248 and BICC. The TDM switch is deconstructed by decoupling the connectivity plane from the central processor and the application software. Connectivity is performed by a packet infrastructure (ATM or IP), and the “rest” of the switch acts as a telephony server. Telephony server solutions seamlessly provision conventional telephony over broadband multi-service networks, reusing the telephony application software and processors. IP telephony The early successes of IP telephony were based on cost and pricing advantages derived from uneven regulations, rather than inherent efficiencies in technology. Nevertheless, IP telephony will continue to successively expand its share of the telephony service market. Increasing investment in the development of IP technology and services will enable the new, multimedia based IP telephony services to mature and benefit from economy of scale. In turn this will permit service providers to move to multi-service network platforms and offer attractively priced separate or packaged IP telephony services. The IP telephony solutions currently available on the market are not conventional, nor do they exemplify the IP telephony of the future. They merely hint at the power of IP telephony for rapid service development and to help operators to quickly and cost effectively move into new markets and businesses. Conversational multimedia services Conversational multimedia services will be the defining feature of broadband multi-service networks and the Mobile Internet. Business users will be the main drivers. They already work in network-based ways and need to increase the effectiveness of their people-to-people communications. Moreover the Mobile Internet industry is person oriented, so its user base will also be important. Until now, H323 has been the platform for IP telephony solutions. It was developed to replace fixed telephony and the standardization was mainly impelled by the relationship with PSTN. Solutions based on H323 have delivered very stable, carrier class IP telephony, and continue to do so. The main stimulus for IP telephony has been its ability to deliver cheap calls by toll bypass – the difference in cost and price created a market window. But margins and prices are decreasing, the cheap telephony window is closing, and the window for multimedia services is opening instead, enabling new revenue streams. The multimedia trend is also strong in the mobile world, including the Mobile Internet, and the convergence of fixed and mobile networks has been anticipated by many analysts for some time. SIP is an emerging Internet standard that allows the flexible integration of messaging, presence and multimedia conferencing with real-time communications such as telephony. It was designed to be modular to integrate applications in innovative ways, and extensible to support new technologies.
Page 14 of 21
Broadband Multi-Service Networks
Firstly, SIP is the choice of the Internet community. The simple, text-based format is easy for application developers to use, as opposed to the binary format of H323. This will lead to development and availability of more applications based on SIP.
Secondly, the 3GPP architecture for third generation mobile networks uses SIP for multimedia communication, which enables the provisioning of common multimedia applications to fixed and mobile users.
Lastly, SIP re-uses existing protocols for web applications such as e-mail and web browsing, while H323 must more or less re-invent the way in which these applications should be handled. Thus, in the SIP world, inter-working will be simpler and smoother.
Streaming services Broadband multi-service networks will create new ways to access and distribute streaming services such as broadcast and interactive video, radio programming and music. One example is “Personal TV ” - a flexible, personal selection of channels and programs chosen from a much larger range of channels than those normally available to the cable or satellite subscriber. Other examples are the communication and broadcasting within companies and large corporations. Such services will be among the most demanding in terms of network throughput and traffic loading, thus requiring appropriate features for efficient handling. They will also need powerful but simple tools that will allow users to conveniently select and control channel or program viewing. Application services and access Application Service Providers (ASP) are a new industry that is set to take off and grow – very much enabled by, and in synergy with, broadband multi-service networks. This is an example of a business that some incumbent service providers are interested in expanding to offer application and access service packages to enterprises. Leased lines, frame relay and IP-VPNs Existing leased line and Frame Relay services can be expected to remain for a considerable time yet, because many enterprises will not replace their end equipment just to change connectivity service. Broadband multi-service networks can carry such legacy, site-to-site, connectivity services over ATM. IP-VPNs (Virtual Private Networks) are becoming an attractive alternative. IP-VPNs provide tunnelled connections through IP networks. Their main benefits are the availability of global connectivity and shorter TTS (Time to Service). Additionally, because IP is in the mainstream of investment and development, IP-VPNs offer greater potential for price reductions.
The access technologies
In summary, we know that end-users demand services and bandwidth. However, they are a heterogeneous group and therefore have different requirements. Moreover the partition of end-users into residential and business categories is no longer sufficient to describe the varying demands made by both groups. Residential users may be defined in several ways: for example as a single-family home or a multitenant dwelling. Business users may range from home workers, to small offices, to large corporations. Operators must be able to provide solutions from the network backbone right up to the end-users, thus providing the broadband services that are required – VoDSL, IP telephony, virtual private networks (VPN), video services, managed bandwidth, ‘always-on’ Internet, etc.
Page 15 of 21
Broadband Multi-Service Networks
A number of different technologies are capable of providing broadband access to business and residential subscribers. Each type of technology has a primary field of application, based on the correlation between its respective characteristics, the different circumstances of the service providers, and the communications demographics of the target customers. Nevertheless the different technologies complement each other, enabling service providers to offer their customers a full set of services with sufficient bandwidth. DSL The term Digital Subscriber Line (DSL), embraces a number of technologies for broadband access over existing copper cables. Asymmetric DSL solutions (ADSL), provide wider bandwidth downstream than upstream. Primarily intended for residential use, this technology typically supports transmission rates of up to 8 Mbps downstream, and about 1 Mbps upstream, both depending on distance and the quality of the copper cable. In practice, standard services offer lower speeds for consistency and reach. Symmetric High-speed DSL solutions (SHDSL), are primarily intended for SOHOs and SMEs. The introduction of a global standard (g.shdsl), should fuel a rapid build-out as local loop unbundling opens up for competition in the copper access area. Very high bandwidth DSL (VDSL) is being considered for two applications – residential access and “extended Ethernet reach ”in campus and building networks. Residential users will benefit from a capacity that enables delivery of multiple digital video streams. Voice Over DSL (VoDSL) refers to the technology that transports voice services over a DSL network access by means of ATM cells or IP packets. The functionality is implemented in a voice gateway and a terminal that together support data access and a number of voice lines. As a technology for providing broadband access over existing copper cables, DSL is an obvious alternative for incumbent operators. However, local loop unbundling is now giving copper access to other service providers, so interest in DSL is no longer confined to incumbents only. Cable TV Many people believe that the capacity of TV cables to transmit images and information is far greater than that of telephone lines originally made to carry voice signals only. Similar to copper and fibrebased networks, CATV can be used to handling Internet and multimedia data. These networks have a maximum upstream capacity of 320 Kbs to 10 Mbs and a downstream capacity of 30 to 40 Mbs. However this bandwidth is shared between all the users on the same segment of the CATV network. The limitations of today’s cable access equipment will continue to create difficulties for operators who wish to balance their investments in infrastructure and the potential revenue gains offered by emerging IP-based data, voice, and video services. However, cable TV access is a great opportunity to reach subscribers with value added services, and vendors and operators are developing solutions to overcome the technological limitations. Fiber to multiple tenant buildings and communities State-of-the-art IT and communication facilities are increasingly seen as a competitive factor in housing and city development. By providing a local broadband communication infrastructure, property developers can increase the attractiveness of their property to prospective tenants or buyers; residential as well as SOHOs and SMEs. The communications infrastructure can also be used to carry IT support for housing services such as security, energy management and key handling.
Page 16 of 21
Broadband Multi-Service Networks
Fiber is a competitive alternative method of providing broadband access to the communities or buildings in which private and commercial users reside. Typically, within the buildings, the separate lines to individual apartments or offices are Ethernet over copper or fiber. New, low-cost installation techniques have made fiber economically, particularly when considering the future demand for very high bandwidth services such as video. This new type of local broadband communication systems will also need local service and application platforms, residential gateways, firewalls and other security features, and effective means of interconnection with the wide area network. Fiber to the office In many circumstances, fiber is the most cost efficient way of providing high-speed access to corporations. Fiber systems with SDH/SONET are already widely used in the access networks of incumbent and new access providers. Typical applications are: fiber to the office, to larger enterprise sites; access network feeder and transport in metropolitan areas, providing a platform for the use of other technologies for the subscriber drop. The penetration of fiber systems in access networks will increase because of rapid overall growth in bandwidth demand. The almost limitless bandwidth of fiber will enable service providers to construct a single physical installation and then handle the growth through CPE changes. Wireless broadband access – LMDS Under certain circumstances, fixed broadband radio is a faster and more cost efficient method of providing high-speed access. Wireless point-to-multipoint broadband technologies such as Local Multi-Distribution Systems (LDMS) are a suitable alternative for access providers who want to use broadband business services to enter new markets, and often for mobile operators seeking to expand into fixed access business services. LMDS is capable of delivering on-demand user data rates of up to 30 Mbps per customer, so it is particularly suited to serve SMEs with broadband access for packet based voice and data services. LDMS offers rapid service rollout, with scalability and flexibility in build-out adapted to customer location and demand.
The packet backbone network
A packet-switched backbone that can handle and scale a wide variety of services is a key component in any broadband multi-service network. It will carry different types of traffic that have diverse characteristics, e.g.: real-time traffic for person-to-person communication; Internet traffic; enterprise communications; video and voice streaming.
Such backbone networks must deliver true, carrier-class resilience and serviceability, and support powerful IP border routing for peering with other IP networks. They must be able migrate gracefully, without degrading the existing services. Advanced quality of service and network engineering features must also be supported, and the network must operate with a comprehensive, unifying management architecture.
Page 17 of 21
Broadband Multi-Service Networks
This type of network consists of a packet backbone running on top of transmission facilities. Current transmission networks typically use SDH/SONET over fiber, but the change to ATM and IP will increasingly remove the need for SDH/SONET. Optical networking over fiber systems is the physical transport foundation of broadband networks. Bandwidth supply has developed at a faster pace than that predicted by the celebrated Moore’s law for micro-electronics. This is made possible by advances in Dense Wavelength Division Multiplexing (DWDM) technology, which increases capacity by transmitting multiple “colors ”, or wavelengths of light, over a single fiber. Advances in IP technology will permit cost savings by running IP directly on optical systems. But, at present, optical line interface standards are insufficiently defined to guarantee the interoperability of line systems from different vendors. To allow a multi-vendor mix of routers and optical systems, Ericsson considers it essential to support open interfaces between the optical and IP layer equipment and, in future, more tightly integrated implementations.
The evolution of new business models places new demands on operators. For example the users of multimedia services delivered over ‘always on’ broadband connections pay for service content and quality rather than call duration. So operators are obliged to manage service levels as well as customer relationships. A management solution for a multi-service network must improve profitability by automating the management processes. It must also minimize fault response times and optimize network performance, ensuring that subscribers receive the service quality that they demand. Performance monitoring also supports the automated optimization of network resources, which can maximize the use of invested capital. To help operators control their costs and stay at the forefront of technological evolution, an advanced management solution is necessary for the development of network strategies that will lead the competition. In a highly competitive and constantly changing service and technological environment, a balance must be struck between proprietary and customized management solutions. Proprietary solutions are quick to implement but inflexible: customized solutions exactly match the operator’s requirements but are slow and costly to implement. To ensure the rapid, profitable rollout of services like Fast Internet Access, IP-VPNs and VoDSL, an automated service provisioning solution is necessary to make the business case successful. Thus the traditional choice – cut costs or increase revenues – is no longer relevant. With an efficient Service Provisioning Solution an operator can achieve increased revenues and cost reductions at the same time.
New requirements on services Due to the dynamics of telecoms industry business models, it is evident that operators are increasingly streamlining their operations and moving towards the outsourcing of all non-core activities. Service providers are chasing turnkey network roll-out projects in which minimum time to first revenue is assured. To succeed in the new telecoms world, the vendors who have strong tools, processes
Page 18 of 21
Broadband Multi-Service Networks
and skilled resources (competence development) to manage the networks and keep them running securely (non-stop services) are those called upon to work with the operators. New business models are evolving, and they place fresh demands on operators. For example, users of multimedia services delivered over ‘always on’ broadband connections pay for service content and quality, not call duration. To support this, operators need to manage customer service levels as well as customer relationships. Service hierarchy Operator needs can also be seen in terms of a hierarchy that is equally relevant irrespective of whether a technological or strategic business dimension is being examined. This “needs” hierarchy has three levels: The first level can be defined as the “basic need”. Essentially it requires suppliers to keep their promises and measure up to expectations. The second level may be called the “tactical need.” It involves how well a supplier can address a customer’s existing business – for example the customer base, and – via the supplier’s expertise and solutions – help the customer to make the most of its present situation. The final level, which may be called the “strategic need,” deals with how well a supplier can help a customer to define the future direction of its business. Today, few vendors can attempt to serve the second and third levels of the hierarchy. To do so, a vendor must have very high technical and business expertise, a complete services portfolio, a large organisation, and often a global presence also.
ENGINE multi-service networks
Ericsson has a clear vision of the network of the future. It envisages a new type of robust, broadband, multi-service infrastructure – based on new packet-switching and routing technologies and designed for real-time services. It supports the business models of operator who have embraced the new world of Mobile and Fixed Broadband Internet. We call it ENGINE. Operators do not invest in Ericsson’s ENGINE concept merely for its advantages as a superior network solution. Of equal importance is the choice of Ericsson as a competent business and network development partner. By working with Ericsson, a network operator gains access to the global knowledge of one of world’s most experienced players, who is committed to the operator’s success. Because Ericsson’s focus, indeed our core business, is to be the preferred partner and supplier to network operators. ENGINE’s value lies in a combination of three promises: • • • ENGINE is the promise of a complete relationship with the operator – sharing experience, plans and risks to attain common business goals. ENGINE is the promise of a complete solution related to the business needs defined by the operator – including end-to-end system integration and other customer services. ENGINE is the promise of a step-by-step roadmap for the operator – from existing technology, through today’s multi-service networks. and onward to the all-IP networks of the future (dependent upon current technological circumstances). From a portfolio perspective, the ENGINE offering is structured to accommodate evolving market segmentation and the new network infrastructure. It comprises:
Page 19 of 21
Broadband Multi-Service Networks
• • • •
Multi-service backbone. Solutions for IP, ATM and WDM backbone infrastructure – with real time capabilities, traffic engineering and end-to-end management. Voice, data and multimedia communication. Migration solutions for legacy telephony and data services based on IP, ATM and new multimedia solutions. Broadband access. Solutions for residential, Small & Medium Enterprise (SME), and Metropolitan access – utilizing XDSL, fiber, fixed radio and cable TV technologies. Applications and application enabling. Revenue generating application with supporting infrastructure for creation and delivery – based on open APIs and third party development support.
Telecom management. A complete suite of end-to end management, including efficient service provisioning solutions. Customer services. A complete solution that supports every phase of the operator’s business – planning, design, build and operate.
Page 20 of 21
Ericsson is shaping the future of Mobile and Broadband Internet communications through its continuous technology leadership. Providing innovative solutions in more than 140 countries, Ericsson is helping to create the most powerful communication companies in the world.
Ericsson Telecom AB SE-126 25 Stockholm, Sweden www.ericsson.com