Strategies for Sustainable Development of Broadband Services

1 Introduction
In recent years, Internet services have witnessed a visible revival and fast development. New Internet services represented by IM (Instant Messaging) are, more than ever, strongly influencing traditional telecommunication services and operating models. IM makes it possible for people from around the world to communicate with each other via text, audio, video or short message all in real time. It can display the status of online friends and make instant conversation through P2P (Peer to Peer) mode. Similar to traditional voice service, IM is featured with real time, interactivity and easy use; but it is much better than traditional voice service because it is nearly free of charge and has outstanding functions. Nowadays, IM service has made a great impact on traditional telecommunication services and this impact is so significant that a significant amount of traditional telecommunication traffic is being substituted by it. This exponential trend is expected to continue. Wanfang Data reports that the number of online IM users exceeds 5.5 million in China in 2003, among which 58% use IM service at least once a day. The number of online IM users is expected to reach 7.4 million by 2004 and 12 million by 2006. By the end of March 2004, around 290 million users have registered with Tencent QQ. Currently, each day there are over 2 billion instant messages are sent through AOL (American Online), while more than 400 million e-mails are sent through AOL per day. Gartner estimates that IM service will surpass e-mail service, becoming the most significant e-communication tool. The Instant Messaging software will be embedded in over 50% of Internet applications. With the progress of information-based society, EIM (Enterprise Instant Message) will become an effective tool to be used in daily work, further increasing the efficiency and effectiveness of business communication.

  Besides IM, other Internet services also turn away part of traditional voice service revenue to a certain degree. A great deal of information is exchanged on the Internet. VoIP works as a common module to be built in various Internet applications such as E-commerce, E-education and Net-meeting. As for tariff policies, people are getting used to receiving free common services (i.e. real-time audio/video communication), but only pay money for special services.

  Since the telecom industry develops fast, operators can obtain growing revenues from access and IDC services. It is true that Internet application users are mutually independent from application users and operators cannot effectively affect the way that customers use applications, let alone stopping the trend of turning away of traditional service revenue. Thus fixed broadband access network become an
 "unwalled Garden". Generally speaking, the development of Internet service is truly bringing more benefits than harm to operators and equipment manufacturers. However, as Internet services flourish more and more, both will face the risk of being edged out. In order to face the new challenge, operators cannot rely only on their traditional experiences. How to develop next generation telecom services? This is an unprecedented challenge for China’s operators. In 1999 annual report, ITU predicts that an unheard of information revolution is coming, and telecommunication is by some reason at the epicenter. In the near future, we are expected to experience lots of changes:

• Operators will pay more attention to
  "service creation" rather than "network construction". They will also witness a big change of business operation model. Network structure will be further opened, which will make it possible for socialized services. It won’t be surprising if telecom industry is mixed with other industries or they enter into an alliance because the boundaries become vague.
• The competition among operators is no more simple, but involving upper and lower telecom industry chain. Each link in this chain will have its own core competence. A company with a kind of core competence usually will only focus on this particular area.
• In general, telecom service revenue will increase. However, the structure of revenues will be changed greatly. For instance, profit margin will drop down; the percentage of voice and access revenue will decrease. In contrast, the percentage of revenue from value-added services and contents will increase. Equipment makers will have to change their product portfolio to adapt to this new structure.
• People are more concerned about market research, service innovation and service development. And telecommunications industry will step into a stage of "new service, new starting point".
• Internet culture will make great impact on operators with respect to corporate culture, operation ideas and organization structure.
  

2 Telecom-Grade Public Service Platform Provided by Operators
Because it is a new telecom service industry chain, an operator should rely on its own core competence and strive for dominant position. Usually operators will be positioned as a sole network provider or "network provider + service provider (SP)". Personally we believe that they are good at constructing and running telecom-grade public service platform in the area of value-added services (especially broadband applications).

  In the current situation, the most appropriate position for operators must be "network provider + public service platform provider". Later after achieving maturity, operators can move towards other service areas. The public service platform is important in a telecom service industry chain so operators should put great efforts into it in an attempt to strengthen their capability for sustainable development.
This platform supports various services and tariff modes, among which two modes must be paid close attention to:
  (1)The first is "SP convergence mode". Users (individual users in most cases) get peer-to-peer man machine services from SP via this platform. Typical services are VoD or information service; users belonging to one SP can enjoy inter-communication service via this platform. Typical services are Instant Messaging and network games. This mode may be applied to enterprises. Accordingly, the public service platform should provide an interface with enough openness and convenient for SP to access easily. The interface shall include provision of service management protocols, network resources (capability) information and so on. This platform will surely benefit SP because it saves SP lots of money on platform construction and routine operation cost. This "SP convergence mode" will make it very simple for SP to be created, accessed and quit. Finally, more and more SPs or even individuals will be attracted to become business partners.

  (2)The second is "user convergence mode". Users (enterprise users in most cases) directly use telecommunication services via this platform. Man-man services and man-machine services are supported regardless of point-to-point or point-to-multipoint. Normally these users have nothing to do with SP. Typical services are public video-conferencing, network video supervision service and the like. Accordingly, the public service platform should have another interface to provide services via a bearing network for enterprise and individuals for easy use. This will obviously save enterprises lots of money on platform construction and routine operation cost. This "user convergence mode" will converge various enterprise application platforms into the operators’ broadband network.
From the above descriptions of two physical interfaces, it clearly states that the common characteristic of a public service platform is serving both SPs and enterprises. Additionally, it also indicates the purpose for operators to adapt to multi-service environment.

  The obvious advantages or core competence of an operator are design, construction and maintenance of a telecom-grade network with large capacity, which is the same in the case of a public service platform. Owing to standardization, resource-sharing and scale effect, a public service platform will effectively reduce SPs and enterprises expenditures on platform construction and routine operation. The public service platform should be flexible for new service creation. As the service platform is continuously upgraded, more and more new services will be introduced, which means better serving SPs and end users. By doing this, the relationship between operators, SPs and end users is consolidated. It should be noted that when public service platforms are developed on a large scale, it will partly affect the structure of telecom industry and equipment manufacturers. Therefore, standardization is very important. One problem that operators would face is that some matured SPs on a large scale or big enterprises do not want to access a public service platform. In contrast, they prefer to have their own platforms. An efficient strategy will be to adopt competitive tie-in sale policies for telecom resources.

3 Next Generation Service Delivery Platform (SDP)
In recent years, all sorts of service platforms have emerged in an endless stream. People are especially widely concerned about the architecture which should be open and service oriented. Some commercial products now use JIAN and PALARY technologies. We sense a phenomenon that in the present-day world, famous IT companies and telecom equipment manufactures (for example: Microsoft, SUN, IBM, HP, Ericsson, Alcatel and Siemens) are strengthening their research on next generation SDP.
In a case where vertical classification is traditionally adopted, services and applications would be tightly bound with delivery channels. Now most devices are working in a separate environment from others, which results in functions overlaying, not-sharing of resources, different information models and not sufficient interface compatibility. Consequently, it shall extend ROI (Return on Investment) period, and increase the cost for service development and maintenance. In this case, traditional solutions expose their inherent defects at aspects of competence, management and standardization. Because of these defects, the traditional solutions cannot adapt to the development of next generation telecommunication services any longer. The service layer of a network requires horizontal solutions, which means that all functions and resources will be shared. It’s a delight that nowadays traditional vertical classification has gradually been replaced by a horizontal classification value chain, and we believe this replacement will certainly bring a significant change on the supply chain.

  How to provide services on a converged network is another new topic. Previously there was a close relationship between service and type of network. To a certain extent, some services may span over several networks. For instance, many PSTN services (e.g. call waiting, call forward) are adopted by mobile communication network. Reversely, some mobile services (e.g. short message, prepaid service) are now absorbed by PSTN. Since packet switching will be the ultimate trend for network evolution, operators want to provide all services. Therefore, the same service will be provided in a very wide range, regardless of mobile or fixed network and broadband or narrow band network. The next generation services will be developed according to Metcalfe’s Law.

  Nowadays the strategies for a new SDP are clear: standard and open architecture shall be adopted; integrated services shall be provided over fixed, mobile and broadband network; centralized and secure services shall serve different types of networks. The next generation SDP shall function as a binder to combine telecommunication infrastructure and IT infrastructure together. It will shorten the period from an idea up to actually generating revenue. The SDP will also bring benefit for reducing operation & maintenance cost and development risk.

  An SDP will have the following features:

• An SDP provides a complete ecosystem for the rapid development, provisioning, execution, management and billing of value added service. 
• Because the combination with IT infrastructure is perfect, resources such as OSS/BSS/MSS, CRM, AAA and portal server may be fully used.
• An SDP can interwork with current network devices (i.e. SSP, SCP, BRAS), new devices
  (i.e. Softswitch, CSCF) and service platforms (i.e. SMS, MMS).
• An SDP supports the delivery of audio, data and video services and contents in a way that is network and device independent.
• It can provide operators, customers and 3rd-party service developers with network capability, service and contents, as well as service development tools. But the way it provides is uniform and standard.
  

  Frankly speaking, the next generation SDP is still under discovery and research. Up to now, there is no fully agreed consensus on which architectural elements fall within the scope of an SDP. Each manufacturer produces its own SDP including different portfolio of products and components.  In general, an SDP should be seen as a bundle of different functional components like an IT product.
SDP is at the service layer of a network, consisting of SEL (Service Exposure Layer), NAL (Network Abstraction Layer), SEP (Service Execution Platform) and CDP (Content Delivery Platform). SEL provides service capabilities to external 3rd party service providers and enterprises via a Web server, for example, Parlay-X GW. NAL provides standardized interfaces to core network elements (switch, signaling and IN devices) and services. Through the interfaces, network capabilities and service characteristics are abstracted, such as call control capability and user location information. By doing this, NAL will ensure that this access is network, technology and vendor independent. Typical devices may be Parlay GW or SIP GW. SEP provides the deployment and execution environment for broad range of voice and data applications, including various value-added services and telecom application servers. CDP provides multimedia contents to user terminals through a way to access original multiple service systems and some systems with supplementary functions. CDP can support multi-path content access (e.g. video, stream media, MMS), download, delivery, authentication, flow control, Dynamic Resource Management (DRM) and terminal device data management.

  As we know that service is important to distinguish operators, and is the key to pursuing sustainable development. When basic voice service has already been provided, customers desire for next generation services, which will be a new challenge for operators:

• How to establish and maintain partnership with ISP (Internet Service Provider), CSP (Content Service Provider) and ISV (Individual Service Vendor)?
• Which new services and network devices should be provided? Is there enough expandability?
• What kinds of network services and capabilities should be exposed? How to expose? Which standard data format shall be used?
• How to create new services? How to integrate standard IT applications?
• How to adapt to the demands for future service diversification, even unknown services?
  

4 Non-PC Terminal
The number of broadband access users increased very fast during the last two years in China. The question that arises is how to keep a high growing rate for a long term? We believe that one of the bottlenecks, which constrain the steady & fast growth, is the monotonous broadband terminal-PC. It is the PC that raises entry barriers for broadband applications so that the broadband user group is limited. According to a report by CNNIC in July 2004, around 97.5% users choose desktop computers to access the Internet. The user group under the age of 30 accounts for 70.5%, the number reaches 82% if counting the users under the age of 35. Students and technical engineers are main users. The reasons for not surfing the Internet are: 37.7% do not have enough knowledge of computer/network, 21.3% have no terminals available. The above figures tell us that two barriers exist for the development of broadband applications. The first is that PC requires certain knowledge of operation skills; it is expensive and takes long booting time. These factors stop many users who are not good at PC, or unwilling to choose PC for broadband service. The second is that currently broadband applications mainly focus on professionals and youth so that many potential users are kept outside of broadband services.

  Be aware that broadband application is not totally equivalent to surfing the Internet. If PC is the only terminal for Internet business mode, it will be unfavorable for healthy development of broadband service business mode. What we should do is to actively develop non-PC broadband terminals and applications. It will help lower the broadband barrier, widen user group and reserve plenty of energy for further development of broadband services. Typical non-PC broadband terminals are set-top
box + TV, broadband multimedia telephone, broadband information terminal, residential GW, which are used for family amusement, home education, video communication and etc. As the broadband access tariff tends to fall, it is better to adopt binding tariff policies, for example, combining fees for broadband access and special value-added services, which are actualized through broadband application terminals. By doing this, ARPU will be increased or ensured.
 

  "Set-top box + TV" is the most typical broadband non-PC terminal, featuring use of a remote control to easily enjoy broadband home applications. Its basic functions are audio/video on demand and broadcast; its enhanced functions are information service, remote education, interactive game and visual phone service. In this way, it will be possible for broadband services to go out of office and game room, letting general people to enjoy so-called broadband living room culture.

   When several broadband application terminals coexist in a house, there is an increasing demand for residential network and GW (Gateway). The residential GW is a simple, intelligent, standard interface unit with flexibility. It can receive communication signals from different external networks and forward the signals to one device through home network. With residential GW, device interworking is achieved. Usually the residential GW may be divided into access GW and application GW. The using of a residential GW will lay the foundation for the development of new home-use broadband services, such as home assistant, network hardware, remote supervision, remote control of household electronic appliances.

  Broadband application terminals will tend to be wireless and diverse. Among different wireless applications, WLAN (Wireless Local Area Network) is getting more and more popular. WLAN will be the most likely to be included in a typical configuration for broadband application terminals. After a certain period of cultivation, a new market or industry in a large scale will be born-broadband application terminal. The popularity of broadband application terminals and residential GW will be affected by several factors such as technology maturity, end user easy of use, standards and price.

  Currently there are many problems in the field of terminal applications: expensive new-type terminal, needing improvement of product ability for easy use, undiversified products, no commercial use on a large scale. Fixed operators don’t have complete requirements for functions and performance from the view of telecom service functions, operations, management, and end-to-end quality assurance.

  There is no standard interface between terminal and platform, which leads to poor compatibility and interworking. Additionally, it is insensitive to cooperation strategies in terminal development, business mode of terminal promotion, and operation & maintenance problems. In order to make a breakthrough, fixed operators should be active to forge a new telecommunication industry chain; and in this chain try to play a leading role by making matured requirements for terminals; and stipulate corresponding standards, even for the interface between terminal and platform.

5 Conclusion
Internet services are witnessing a visible revival and fast development, which brings more and more influence on traditional telecommunications industry. Both operators and manufacturers should actively transform their position and strategies.

  In the current situation, the most appropriate position for operators must be "network provider + public service platform provider". Equipment manufactures should pay more attention to service research, digging out the requirements of the "customers of customers". They should also follow the track of next generation service delivery platform, and start R&D efforts on telecom-grade public service platform. The monotonous PC terminal raises the entry barrier for broadband services, which constrains the stamina for further development of broadband services. An appropriate strategy is to actively develop non-PC broadband terminals and applications.

Manuscript received: 2004-08-12