Trends in Deployments of New Telecommunications Services by Local
Exchange Carriers in Support of an Advanced National Information
Statement of the Challenge
The telecommunications industry, the computer industry, and other industries have been performing research and development for more than two decades directed toward the realization of "information age" applications of computer and communications technologies. In the last few years, this vision has been articulated in the contexts of the "information superhighway," the "national information infrastructure'' (for the United States), and more broadly as the "global information infrastructure." While the definition of a national information infrastructure (NII) is subject to some differences in viewpoints, several consensus definitions have been published. For example, the definition published by the private sector Council on Competitiveness includes the following paragraph:1
The infrastructure of the 21st century will enable all Americans to accessinformation and communicatewith each other easily, reliably, securely, and cost effectively in anymediumvoice, data, image, orvideoanytime, anywhere. This capability will enhance the productivityof work and lead to dramaticimprovements in social services, education, and entertainment.
Although definitions of the emerging NII may differ in detail, they appear to be quite similar in spirit and intent in their focus on the following features:
A key component of such an NII is the underlying communications fabric, which allows users to connect with other users via their communicating/computing appliances (telephones, computers, personal digital assistants, fax machines, set-top boxes, etc.). In the United States, this underlying communications fabric is composed of the diverse networks of local exchange carriers, cable TV providers, wireless (cellular) providers, alternate access
NOTE: Submitted on behalf of Ameritech Corp., Bell Atlantic Corp., Bellcore, BellSouth Corp., NYNEX Corp., Pacific Telesis Group, SBC Corp., and US WEST Inc.
providers, interexchange (long-distance) carriers, and value-added networks that are built on top of these (e.g., the Internet).
Traditional telecommunications networks have satisfied many of the requirements implied by the vision of the NII, and, indeed, form the communications fabric of today's information infrastructure. They are affordable, ubiquitous, easy to use, and dependable, and they have supported a wide and increasing range of applications including telephony, data communications (using modems), fax, access to the Internet, voice messaging, e-mail messaging, voice-response services, and access to variety of information services. In addition to the applications listed above, which are supported by ubiquitous dial-up telephone services subscribed to by 94 percent of households,2 there is a variety of higher-speed and/or specialized telecommunications services provided to businesses and institutions for such things as high-speed data transport and video teleconferencing, and for interconnecting Internet routers (packet switching nodes).
The ongoing challenges in telecommunications networking today focus on the following:
Meeting these challenges in providing an advanced communications fabric for NII applications requires the investment of billions of dollars of research and development funds, and the investment of hundreds of billions of dollars in new network facilities on a nationwide basis over the next two decades. These investments include the installation of combinations of optical fiber, coaxial cable, wireless technologies, and network software throughout the United States. One cannot overestimate the challenges associated with making networks and network services reliable, secure, and easy to use, and doing so at costs that are compatible with the expectations and ability to pay of residential and small business consumers. The vast majority of these software investments are directed at meeting these challenges. Since the demand of residential and institutional consumers for the newer applications that are envisioned within the framework of the NII is highly uncertain, and by implication the demand and associated revenues for the telecommunications services that the advanced communications platform can support are uncertain, these investments involve high risk, except in situations where a combination of existing revenue streams and cost savings can justify the investments independent of the demand for speculative new services. The rapid depreciation of computer and communications technologies, in terms of rapidly improving performance/price ratios, makes these investments even more risky because investments made in advance of market demand may never be recovered in a competitive marketplace.
Further compounding the risk associated with the large investments required to put in place the telecommunications fabric of the NII is the uncertainty associated with the regulatory and legal framework within which network providers must operate. The regulatory and legal framework of the past is ill suited for an environment of large investments targeted toward highly uncertain market needs using rapidly depreciating technologies in a competitive marketplace. For example, the requirement of a network interface device erects an artificial barrier that prevents local exchange companies from providing complete services to their customers. The regulatory and legal framework of the future is still being defined in a slow-moving set of processes. These
processes are, to a large extent, setting the pace for the deployment of the next-generation telecommunications fabric needed to support the full range of envisioned NII applications.
Background: Evolution of the Networks
Since the creation of the former Bell system and passage of the Communications Act of 1934, and until the past decade, the business of telecommunications has primarily been focused on cost reduction and improving the quality of telephone calls in terms of the ease and speed of setting up a call and the quality of communication after the call is set up. In the late 1940s and early 1950s, capabilities to allow direct dialing of long-distance calls by calling customers were created and deployed on a nationwide basis. In the 1960s and 1970s the emergence of modern electronic circuitry made possible the introduction of microwave and coaxial long-distance cable systems that provided higher-quality connections and achieved lower transmission costs. In this same period, and continuing throughout the 1980s and 1990s, the introduction of computing technology, both in stored program-controlled switching systems and in the automation of network operations to streamline and facilitate manual tasks, resulted in dramatic increases in efficiency and the ability to serve customer needs quickly. The introduction of fiber-optic systems, starting in 1979, further improved the quality of local and long distance connections and further reduced the costs of transmissions. The introduction of local digital switching systems in the 1980s dramatically reduced maintenance costs.
Figure 1 shows a trend in the total number of local exchange access lines per employee in a typical U.S. telephone company over the last two decades. In the past decade, this number has increased from 167 access lines per employee to 250 access lines per employee. This improvement in efficiency has been enabled by the ongoing investment in new network technologies such as fiber optics and software-controlled remote (unattended) electronic systems, and in software-based systems that are used to facilitate all aspects of the business, including negotiating with customers to take orders for new or changed telephone services, determining the availability of equipment that can be assigned to new customers, assigning installation staff to connect customers to equipment that has been reserved for their use, and determining the causes of service problems and arranging repairs. With state-of-the-art computerized systems, which involve tens of millions of lines of code, many of these functions can be substantially or completely automated.
In addition to the cost reductions that have been achieved by the continuous investment in advanced technologies (hardware- and software-based), there has traditionally been an emphasis on the use of subsidies to make basic residential services universally affordable. Business services and long-distance services have traditionally subsidized residential service. Services in locations that have lower associated costs (e.g., urban areas) subsidize services in locations that have higher associated costs (e.g., rural areas).
These subsidies have resulted in low prices for basic residential services. As mentioned, this has resulted in the ubiquitous availability of affordable telephone services. However, in many places in the United States, basic telephone service is priced substantially below cost. As the nation moves into an environment of
competitive provision of new telecommunications networking services, the historical subsidies will slow down the widespread deployment of new services for the following reasons:
In the last 10 years, the traditional focus on reducing costs and improving the quality of telephone service has been supplemented with a focus on providing new telecommunications services that are associated with the vision of the emerging information age, as captured recently by the vision of the NII. Three of the main thrusts that have emerged in the context of services directed toward meeting the needs of mass markets (residential and small business customers) are as follows:
Conversion of Networks from Analog to Digital Technologies
Over the last several decades, starting in 1962, digital transmission technologies have been introduced in public telecommunications networks. Initially, this took the form of T1 carrier service on relatively short-distance connections between telephone switching machines. Since 1979, it has taken the form of fiber-optic systems that link switching systems and that reach out directly to business customers, and to unattended remote terminals that serve residential customers. In addition, since the second half of the 1970s, analog switching systems have been upgraded to newer systems that employ digital switching technologies. The net result is the ability to provide end-to-end digital services to customers to support emerging multimedia applications.
The Advanced Intelligent Network
The introduction of stored program-controlled (computerized) switching systems starting in the 1960s made it possible to go beyond direct dialing of calls to offer customized telecommunications services to meet users' needs.
The earliest customized services to be offered included customer-controlled call forwarding, three-way calling, and network-based speed dialing. Recently these have been supplemented by services that depend upon the calling party's number such as caller identification, automatic recalling of the last call received, and call blocking. These services depend upon the combination of low-cost memory (storage of information) and processing power that is enabled by state-of-the-art electronic technologies. However, these types of services have traditionally been implemented by making changes and additions to the large software programs that run the switching systems. Making changes to these large mainframe-like systems is very costly and time consuming. Furthermore, since switches are purchased from a multiplicity of suppliers and come in a multiplicity of types, implementing new services has traditionally required the development and deployment of new generic switching software by multiple suppliers for multiple switch types. This costly and time-consuming process is not consistent with the rapid deployment of a wide range of new telecommunications services that are customized to meet users' needs.
Thus, the local exchange carriers have implemented the AIN as a client-server approach to creating new services. In this approach, the switches act as clients that interface with software-based functionality in server nodes called service control points (SCPs), service nodes, and intelligent peripherals. The switches, service nodes, and intelligent peripherals implement building block capabilities that can be mixed and matched by the SCPs to create new services for network users. Since all switches implement comparable building block capabilities, new services can be created and deployed quickly by implementing new functionality in the server nodes. Following are three examples of new services that can be implemented in this way:
through a sequence of numbers that may depend on the time of day, day of week, or other parameters chosen by the subscriber.
The key advantages of the AIN are that it can provide customized telecommunications services to end users in a manner that is easy to use and that works with a wide variety of appliances ranging from simple telephones to advanced personal digital assistants.
Advanced Digital Access
Today's residential and small business telephone customers typically access the network using dial-up telephone services delivered over the copper wires that connect them to the network (except wireless customers as described below). Using modern electronics, it is possible to extend the range of uses of these wire-pair-based connections far beyond what was originally contemplated. For example, the latest modems allow for digital communication at 28.8 kbps on dial-up connections. In order to support multimedia applications such as access to stored images, video clips, compact-disk-quality audio clips, and two-way multimedia teleconferencing/collaborative work, customers need more than a modem and a dial-up line. One of the largest technical, economic, and regulatory challenges facing the telecommunications industry is how to create a path forward toward a ubiquitously available, affordable (by residential users), high-speed digital capability to support multimedia applications.
One of the steps in this direction is the widespread deployment and availability of integrated services digital network (ISDN) capabilities. In most cases, ISDN uses existing telephone wires to provide two-way, simultaneous digital connectivity at up to 128 kbps, plus a 16-kbps digital channel for network control signaling and additional packet data. Users of ISDN have reported greatly facilitated access to the Internet World Wide Web and other sources of multimedia information, and the ability to carry out multimedia teleconferencing/collaborative applications that include face-to-face video that is of reasonably high quality.
Because ISDN uses the existing wire pairs for most users (the distance from the user to the terminating point in the network is the key factor, with roughly 6 km being the limit using existing wire pairs), it can be provided with a relatively moderate (but still large) initial capital investment. The existing switches must be upgraded with hardware and software to support ISDN. Analog switches must be replaced with digital switches, or the ISDN customer must be reterminated on a digital switch. Note that approximately half of the telephone lines today terminate on stored program (computer)-controlled analog switches that still offer high-quality service for wire-pair telephone lines.
The software-based operations support systems that are used to automate operations in the network must be upgraded to accommodate ISDN. The individual customers who order ISDN must employ special terminations that are compatible with the ISDN interface, whether built into an ISDN telephone, a special interface board of a computer, or an ISDN terminal adaptor. The decision of a RBOC or a local exchange carrier (LEC) to make ISDN available throughout its territory (make it available to its roughly 10 million to 15 million subscribers) is a multibillion-dollar investment decision. The capital investment by a customer for an ISDN interface is a commitment of several hundred dollars at this time, but this cost will drop rapidly as ISDN usage rises over the next several years.
The ongoing charges for ISDN access vary throughout the country and are based on a combination of a flat monthly fee and a usage charge that may depend on minutes of use, packets sent, or a combination of these. As mentioned above, the challenge for the telecommunications provider is to recover investment costs in an environment of traditionally subsidized, below-cost pricing of basic residential telephone services.
Beyond ISDN, in the intermediate and longer term, is the challenge of providing residence and small businesses with digital access that is capable of supporting applications such as high-quality, full-motion video. Such applications require more than 1 Mbps and can range up to 20 Mbps or more for full-quality compressed high-definition television signals. To provide such services requires the deployment of a new physical infrastructure consisting of optical fibers, coaxial cable, and broadband wireless technologies. The investment cost is likely to be in the range of $1,000 to $2,000 per subscriber location served, amounting to several hundred billion dollars on a
national level. Since the applications and the user demand are speculative at this time, creating an economic, regulatory, and legal framework that will encourage investors to take the enormous risks associated with such a deployment is a national challenge. Network providers are reluctant to deploy higher-cost architectures and technologies 5 or more years in advance of speculative market needs for such things as very high bandwidth (more than a few hundred kilobits per second) upstream capabilities on a per-user basis. Nonetheless, some RBOCs have announced plans to deploy broadband networks throughout their service areas over the next 20 years, and all RBOCs are deploying broadband networks on a limited basis to test market demand and technology capabilities.
In a competitive marketplace, other firms can learn from mistakes made by the first entrant and can then enter the marketplace later with newer technologies and a better understanding of customer needs. The potential advantages gained by waiting may give rise to a "getting started" problem, as all potential investors wait for someone else to make the first move. Long-term, large-scale projects like the NII may not be allocated adequate capital. To offset this risk partially, network providers would like to begin recovering their capital investments as soon as these new networks are deployed. Existing revenue streams for traditional telephony and entertainment video services are less risky than unproven and speculative new services. By offering old and new services on a new shared platform, network providers can reduce their revenue risk and also benefit from economies of scope. The need to reduce risk and share the costs of network deployment across many users and services may be an important factor driving many telecommunications companies' interest in entertainment video and many CATV companies' interest in traditional telephony.
Services to Support People on the Move
Since its introduction in 1984, cellular telephony has grown approximately 40 percent per year to serve nearly 20 million customers today. Paging services and cordless telephones are also highly popular. In the next several years, new kinds of personal communications services based on digital technology and supporting both traditional voice and computer/multimedia applications are expected to be widely available. It has been estimated that by the year 2003, there will be 167 million U.S. subscriptions to personal communications services, with many customers subscribing to multiple services.3
While wireless provides the physical access mechanism for an untethered telephone and other appliances, the advanced intelligent network (AIN) provides the software-based functionality to people on the move. Home and visitor location registers (AIN service control points) keep track of where nomadic users are and provide the information required to direct incoming calls to those users. AIN can screen or block incoming calls according to the calling number, time of day, day of week, or other parameters specified by the called party. AIN functionality allows "multitier" telephones to access cordless base stations, high-power vehicular cellular base stations, low-power pedestrian cellular base stations, and, in the next several years, low-Earth-orbiting (LEO) satellite systems, depending on which is most economical and available at any given time. As wireless telephony transitions toward nomadic multimedia computing and communications, the advanced intelligent network will provide access control (security-related) mechanisms, interworking functionality, screening, customized routing, media conversion, and other "middleware" functionality to support people on the move.
Advanced Intelligent Network
Based on demographics, it is probable that all RBOCs and most other local exchange carriers in the United States will deploy the advanced intelligent networking capabilities described above nearly ubiquitously over the next 5 to 7 years. Some carriers have already made these services widely available. These services will represent to telecommunications subscribers what the advent of the personal computer represented to its user community. Users will be able to define and obtain customized call processing capabilities to support both voice and data/multimedia applications such as customized screening and routing of calls, automated media conversion to facilitate the delivery
of messages, personalized telephone/network numbers, and access control (security-related) services. These will be provided in a way that is easy to use, reliable, affordable, and capable of interworking with a wide variety of appliances and terminals, ranging from simple telephones to personal digital assistants. AIN will enhance multimedia communications by enabling users to control multiple channels in a single communications session, and by interfacing with a variety of user terminal devices in a user-friendly way.
Integrated Services Digital Network
ISDN is widely deployed and available today. A detailed deployment schedule for ISDN is shown in Figure 2. ISDN is a major step forward in enabling two-way, interactive access to multimedia information, multimedia messaging, and multimedia teleconferencing and collaborative work. It will be the backbone of the transition of residential access toward broadband over the next 20 years. Along with today's dial-up modem-based access, ISDN will be a principal access technology for residential and small business users accessing the Internet over the next 20 years. ISDN, in both its basic rate (two 64-kbps "B" channels) and primary rate (twenty-three 64-kbps "B" channels) forms, will be used by businesses to meet their traditional telecommunications and Internet access needs, and it will be used by cellular and emerging personal communication service (PCS) providers to connect into the core telecommunications networks. ISDN will be a principal access mechanism for K-12 schools, libraries, and community centers to connect to the national information infrastructure.
Higher-Speed Switched and Nonswitched Services
Until recently, the primary method by which businesses and institutions obtained nonswitched private line connections between their locations was to use dedicated 1.5-Mbps T1 lines, and dedicated 56-kbps digital private lines rented from telecommunications carriers, including the local exchange carriers. Some larger businesses and institutions have used higher-speed 45-Mbps private lines for point-to-point connections. Recently, new types of digital services, including frame relay, switched multimegabit data service (SMDS), and ATM cell-relay, have been introduced by telecommunications carriers, including local exchange carriers. SMDS is a packet-switched, connectionless data service that allows the destination to be specified independently for each packet. Frame relay and ATM are currently nonswitched services that utilize predetermined destinations for traffic; switched versions of these services are under development. All these services offer the advantages of improved sharing of facilities (fibers, terminations on electronic equipment, etc.) through statistical multiplexing. These new services, particularly ATM, can also support advanced multimedia applications that require high data rates and low delay variability between communicating endpoints.
These higher-speed services are being deployed in concert with market demands and are expected to be widely deployed and available over the next 5 to 7 years.
Cellular networks are widely deployed in urban and suburban population centers, and coverage and connectivity are steadily improving. These networks are being expanded to meet the growing user base with the deployment of smaller cells and newer technologies. Low-power cellular (personal communications services) to support people on the move is being implemented and will be widely deployed over the next 5 to 7 years. Wireless networks are being upgraded to employ digital technologies that support data and multimedia applications. In addition, these digital technologies enable the incorporation of encryption methods to improve the resistance of wireless services to eavesdropping. The use of advanced intelligent network functionality and services will allow for improved roaming and mobility for wireless users and will enable access to multiple wireless networking
services (e.g., cordless telephony, high-power cellular, low-power cellular, and satellite-based services) from a single telephone handset. Such "multitier" applications are being deployed now by some local exchange carriers, and they are expected to be widely available over the next 5 to 7 years.
The Internet, as it exists today, is built on services provided by local exchange carriers and interexchange (long-distance) carriers. Users access Internet routers (switches) through dial-up telephone lines and 56-kbps or 1.5-Mbps T1 private lines leased from telecommunications network carriers, primarily local exchange carriers. Routers are interconnected with 56-kbps, T1, and 45-Mbps private lines, typically leased from telecommunications carriers. Increasingly, fast packet services (such as frame relay and SMDS) are being used to replace point-to-point links.
Recently, several local exchange carriers have announced offerings of complete Internet Protocol (TCP/IP) offerings, including routing functionality, mail boxes, and support services. It is likely that most local exchange carriers will offer complete Internet service product lines in the next several years. However, there are regulatory issues that can delay the RBOCs' offerings of Internet services. The Modified Final Judgment (MFJ) prohibits the RBOCs from carrying traffic that crosses local access and transport area (LATA) boundaries; such traffic must be handed off to a long distance carrier selected by the consumer. It is not clear whether, and if so, how, the restriction applies to the provision of Internet service. In testimony before the House Subcommittee on Science, George Clapp, general manager of Ameritech Advanced Data Services, made the following statement:
Offering a ubiquitous Internet access service with the burden of the longdistance restriction wouldincrease our capital costs by 75 percent and expenses by 100 percent. Thefollowing factors contribute tothese additional costs:
At this point, some RBOCs are interpreting the MFJ restrictions to apply to their Internet service offerings. This is an area where regulations need to be changed to allow the RBOCs to compete on an equal basis with other carriers that are not subject to MFJ restrictions.
As described above, the provision of ubiquitous, affordable broadband access to residences is one of the most difficult challenges facing telecommunications carriers. All RBOCs have expressed a commitment to deploy broadband access services as quickly as the market demand, technology cost trends, and regulatory and legal environment permit.
The RBOCs have collectively invested approximately $20 billion per year in upgrades to their networks since divestiture in 1984. They have committed to increase their investments substantially if regulatory reforms are enacted that enable them to be full-service providers of telephony, video, and multimedia interactive services in an environment that is conducive to the high-risk investments required to deploy broadband access networks. Several of the RBOCs and other local exchange carriers have market trials of broadband access under way or planned.
The deployment of symmetrical two-way capabilities, which permit residential users to originate individualized very high speed (greater than several hundred kilobits per second) upstream communications is a major challenge. One must differentiate between the concept of symmetrical two-way access, which has been raised as an issue by the government and other stakeholders, and the concept of two-way capability. The most demanding two-way capability that has been identified in the context of networks that serve residences is two-way multimedia collaborative work, also called multimedia teleconferencing. Research has shown that two-way multimedia collaborative work can be supported, to a large extent, by basic rate ISDN, and that nearly all needs can be met with a two-way capability of 256 to 384 kbps. Most broadband access network architectures being considered for deployment by the RBOCs can support this capability on a switched basis for all subscribers. At issue is whether there is demand for still higher speed two-way capabilities, comparable in speed to the one-way capability needed to deliver entertainment-quality video to residential customers. The data rate associated with entertainment video ranges from 1 Mbps for VHS quality to 20 Mbps or more for HDTV quality video. The ability to deliver entertainment-quality video both downstream to residential users as well as upstream from residential users is what is called symmetrical two-way access.
Although a large number of alternative architectures have been extensively studied from a capability and cost perspective, it appears that in many situations substantial incremental investments are required to provide symmetrical two-way capabilities. It is unlikely that these incremental investment costs will be recovered in a competitive marketplace if they are made many years ahead of the demand for such high-speed upstream services. The details of the trade-offs among alternative broadband architectures vary from RBOC to RBOC depending on such things as the density of housing units.
Dependable, Usable Networks
The tradition of the telecommunications industry has been to provide network services that are highly reliable, secure, and usable by the widest possible range of telecommunications services customers. As new, interactive, multimedia networking services and applications are deployed, using a wide range of new and
heterogeneous technologies, it will be a great challenge for all industry participants to maintain this tradition in the context of the NII. If individuals, corporations, and institutions are to reengineer themselves to become dependent on networked applications, then those individuals, corporations, and institutions must be provided with network-based services and applications that are even more dependable than today's telephony services. They will expect those services and applications to be easy to use, to work all of the time, and to be secure from intrusions and other security threats. The RBOCs are committed to maintaining their tradition of reliable, secure, and easy-to-use services through a combination of technological and operational methods. In particular, the use and sharing (in public forums such as the National Security Telecommunications Advisory Committee) of best practices among network providers are essential to help prevent and minimize such threats. Cooperative testing between networks to detect incompatibilities, particularly of management protocols that protect faults from propagating into large outages, is an essential ingredient of this process.
As we move into the future, the role of telecommunications networks in facilitating interoperability and ease of use will become increasingly important to consumers. While early adopters and those who create new technologies have a relatively high tolerance for complexity and unreliability and are willing and able to invest substantial amounts of time in learning to use applications and in resolving problems, mass market users expect their applications and services to be extremely dependable and intuitive. In theory, software-based functionality can be placed in end users' terminals to enable interoperability, to resolve incompatibilities that would be perceived by customers as application failures, and to make complexity transparent to end users. In reality, this is achieved today by forcing end users to be systems administrators of their complex terminal software, or to engage others to administer their systems for them. Traditionally, the telephone networks have hidden complexity from end users and have resolved incompatibilities among end user terminals by employing ''middleware" in the networks. For example, an end user in New York can make a call from an ISDN telephone to an analog cellular phone in London. As applications such as multimedia teleconferencing, multimedia messaging, and remote access to multimedia information become increasingly important in mass market applications, telecommunications networks will play a critical role in resolving incompatibilities between different types of user terminals and between user terminals and servers, in facilitating the location of resources, in helping users manage their communications services, and in providing capabilities such as multimedia bridging.
Most of the technology-related challenges in creating the national information infrastructure can be best addressed by the private sector, with the cooperation of the public sector.
1. Council on Competitiveness. 1993. Vision for a 21st Century Information Infrastructure. Council on Competitiveness, Washington, D.C., May.
2. Federal Communications Commission. 1994. Statistics of Communications Common Carriers, 1993/1994 Edition. Federal Communications Commission, Washington, D.C., Table 8.1.
3. Personal Communications Industry Association. 1994. 1994 PCS Market Demand Forecast. Personal Communications Industry Association, Washington, D.C., January.
4. Clapp, George H. 1994. Internet Access. Testimony before House Subcommittee on Science, October 4.