(Approved by the IEEE-USA
Board of Directors, 15 Feb. 2001)

The Institute of Electrical and Electronics Engineers - United States of America (IEEE-USA) has considered what government regulatory posture will best foster the development of the U.S. telecommunications infrastructure over the next decade to realize the greatest economic and social benefits given foreseeable technological advances.

IEEE-USA recommends that Federal and State regulators:

• Promote technology-based competition. That is, allow all technologies to participate in any market segment.

• Exercise technologically neutral regulation. That is, no regulatory provision other than spectrum availability should constrain any one technology in developing its full capability.

• Require open access to each technology by application and content providers. That is, the broadband facilities should constitute a level playing field for diverse services and information.

• Encourage voluntary development and adoption of open industry standards within each technology. That is, the user of a particular type of broadband facility should encounter the similar specifications and performance from any supplier at any stage of development.

• Encourage universal access to these technologies.

These principles will allow suppliers of both carriage and content to compete on technological merit in serving their highest applications and markets. If one technology approaches a dominant position in a given market, regulatory oversight should then be exercised to foster further development and operation in the public interest.

IEEE-USA’s Committee on Communications Policy reviewed the present and future capabilities of four promising broadband technologies, as extensively reported in the April 2000 issue of the journal Info(1) and summarized below. These technologies, now being actively built out by various facilities providers, are: optical fiber, hybrid fiber coax, digital subscriber line, and both terrestrial and satellite wireless. The conclusions are:

• At present all four broadband technologies can provide useful and economically viable multimedia services such as voice telephony, data transmission (including Internet access), music, and video.

• In the future, all four technologies will compete for existing and expanded markets on the basis of performance and cost. Thus, although it is possible to estimate future capabilities, only further development and deployment will determine how they will ultimately share in delivering various applications to particular markets.

• The greatest transmission capacity and switching flexibility at lowest marginal cost probably lies with a predominantly optical fiber plant all the way to the end-user's premises.

• Hybrid fiber coax has impressive broadband capabilities (say, 1 Gbps) but suffers from a shared-use architecture, which may result in congestion during peak loads.

• Digital subscriber line is a near-term solution using existing copper plant, but will reach distance-dependent transmission-rate limits that are much below optical fiber and hybrid fiber coax.

• In the course of development, hybrid fiber coax and digital subscriber line will necessarily extend optical fiber trunks out from the head end or central office, ultimately approaching much nearer to the end-user's premises than current practice.

• Fixed wireless can provide reasonably high transmission rates up to physical limits imposed by radio propagation phenomena. Mobile wireless, of course, is essential for mobile applications.

This statement was developed by the Committee on Communications Policy of IEEE-USA and represents the considered judgment of a group of U.S. IEEE members with expertise in the subject field. IEEE-USA promotes the careers and public policy interests of the nearly 230,000 electrical, electronics, computer and software engineers who are U.S. members of the IEEE.

BACKGROUND

The issue is how best to foster the development of the U.S. telecommunications infrastructure over the next decade to realize the greatest economic and social benefits, given foreseeable technological advances.

Importance of a Future Broadband Infrastructure

The role of broadband, digital telecommunications services to the end user will become critical to the U.S. economy in the next few years. Businesses are using it widely now and residences are on the threshold of adoption. It already facilitates electronic commerce from business-to-business, and business-to-consumer demand is growing. Globally it advances the competitiveness of U.S. industry and trade. Domestically it will provide easy and inexpensive access to health information, online education, government services, and entertainment, thus raising the quality of life for our people. Companies are aggressively building out plant using the technologies of optical fiber, hybrid fiber coax, digital subscriber line, and wireless, discussed below.

Forecast Capabilities of Promising Technologies

This section summarizes the technologies and associated architectures that may be expected by 2010, as described in the committee’s published work(2) While each paper in that work primarily discussed one technology, it also considered competing technologies and related issues, thus resulting in some overlap. This summary consolidates the discussion of each technology under one heading, and avoids overlap.

Optical Fiber Technology

The major advantage of optical fiber is its high capacity: a backbone trunk operating at 40 Gbps and using Dense Wave Division Multiplexing (DWDM) allows transmission rates approaching terabits per second or higher. Currently, fiber is used extensively in conjunction with coax cable to provide high date rate services (e.g., interactive video, video on demand, multimedia, and high definition television). However, the architecture may migrate to all fiber as a result of increasing number of subscribers and greater demand for higher data rates.

An all fiber network (AFN) would bring fiber to the home or the building (FTTH or FTTB). It may initially use passive optical splitters at the far end, which are inexpensive and do not require costly adjustments or power supplies. The transmission on such a network is asymmetric: the "reverse" path (from user to head end) cannot carry the same high data rate as the "forward" path (from head end to user). Later on, passive splitters may be replaced by fiber-based active local area networks (LAN), using 10Base FL/100Base FX equipment. This architecture allows software controlled upgrades, and can bring fiber to the home as lower cost light sources and connectors become available. However, use of active components increases maintenance cost; and the installation is more expensive because the required Ethernet switch needs an environment-controlled housing, and the active elements need a power supply and back-up power.

DWDM allows different services to be put on different wavelengths (e.g., analog video and digital video can be carried on separate wavelengths), and new services can be added on previously unused wavelengths without affecting other services or requiring new equipment. In future, it may be possible to dynamically route different wavelengths, thus providing multiple virtual fibers to the home, with very large bandwidth.

A major technical challenge for AFN is the choice of signal protocols. Internet Protocol (IP) does not allow assignment of priority for time-sensitive traffic such as voice or video; it also suffers from latency (if some packets have to be retransmitted several times due to errors), which may not be acceptable for voice or video. These drawbacks can be overcome by repackaging IP packets into Asynchronous Transmission Mode (ATM) or Synchronous Optical Network (SONET) envelopes. However, this increases the cost significantly and reduces the available bandwidth due to repackaging overhead (e.g., 6-fold increase of cost and 11 percent reduction of bandwidth for ATM). Another challenge is to develop low-cost reliable power supplies, e.g., lithium-ion batteries or fuel cells.

Although the installation of fiber networks lags that of other systems, in the long run fiber may prevail because of its advantages of bandwidth, flexibility, robustness and manageability over competing technologies.

Hybrid Fiber Coax (HFC) Technology

Coaxial cable television systems serve nearly 97 percent of television households in the United States. They operate at frequencies from a few MHz to more than 1 GHz, the upper limit (about 2 GHz) depending on the dielectric material used. Typical coax bandwidth is 750 MHz, containing 112 channels of 6 MHz each in a total spectrum of 672 MHz. Because of high attenuation along the cable, amplifiers are needed every few hundred feet, or a maximum of one or two thousand feet, apart. Typically a tree and branch architecture is used to serve large clusters of 500 to 2000 homes. This architecture provides the same signal to all customers and makes it difficult to offer them individually switched services.

An HFC system uses fiber from a Central Office or head end to an Optical Node (ON), where the signal is converted to radio frequency and carried by coax cable to a smaller cluster of (say 200) homes. Each area served by the ON can be further divided into smaller clusters, and the same spectrum can be reused several times to provide different services. Thus, MPEG video, analog video, switched data, and IP data can be carried as several "virtual" networks on a single physical network.

The ONs can be brought closer to the homes or buildings as more customers demand higher data rates, ultimately resulting in an all fiber network (AFN). Further, the architecture can be converted to a ring by interconnecting the ONs through optical fiber. This architecture allows the provision of different services to individual clusters or sub-clusters, and offers a high degree of flexibility in segmenting the service area.

Future HFC networks will be able to provide many "common" signals (such as analog, digital, and high definition television), and several "dedicated" or specialized services (such as telephony, IP data at standard and high speeds, and video on demand). They can also leave several channels free to permit introduction of new, as yet undefined, services.

Digital Subscriber Line (DSL) Technology

Digital Subscriber Lines use unshielded twisted copper wires of legacy telephone networks to provide high data rate services to homes and businesses. Typically, the service is asymmetric (ADSL), with 1.5 Mbps in the forward or downstream direction (to customer) and 640 kbps in the reverse or upstream direction (from customer). Other variations include High Speed, ISDN, and Very High Speed services, called HDSL, IDSL, and VDSL, respectively. Their speeds vary between 64 kbps to 52 Mbps downstream, and 64 kbps to 6 Mbps upstream. Customers are provided dedicated circuits, with inherent benefit of security and being always available when needed, even when the power line is down. (However, some companies also offer dial-up DSL at lower cost.)

DSL was deployed rapidly because excess telephone capacity was available. However, unshielded twisted copper wire has a major shortcoming: it cannot support high data rates over long distances from the Central Office (CO). Typically, 1.5 Mbps can be carried out to 18,000 feet from the CO, 6 Mbps to 11,000 feet, and 8 Mbps to 5,000 feet. To extend the service area, a hybrid DSL-Fiber architecture is used, similar to HFC. Thus, fiber is used from the CO to a remote location, and copper wires extend from there to the customer. Ultimately, this may lead to an all fiber network (AFN).

Another shortcoming of DSL is cross talk between circuits; it becomes a serious problem if the proportion of copper wire packaged in a bundle reaches 40 percent and is even more severe at higher data rates. (This is why DSL penetration is currently limited to 70 percent of all homes in the United States, and why most service offerings are limited to 6

Mbps.) Use of hybrid DSL-fiber architecture partially mitigates the cross-talk problem, thus allowing higher user densities and higher data rates. Some DSL providers are also considering a hybrid DSL-wireless architecture, since wireless technology is making rapid advances and may offer low-cost solutions to the "last mile" (or rather the "last step") problem.

Wireless Technologies

Wireless can provide service to fixed and mobile users through terrestrial or satellite networks. The major advantages of wireless technology are its ability for rapid deployment, reducing the cost of infrastructure, serving sparsely populated areas, and providing global access. Limited development is also being done on wireless systems using unmanned aircraft.

Terrestrial Systems

Terrestrial Fixed Wireless Access (FWA) systems currently operating in the microwave and millimeter bands offer duplex (simultaneous two-way) data rates of up to 155 Mbps, and simplex (one way) services.

Most of the technical evolution in the next decade for direct-to-user, two-way fixed wireless services will take place in the 20-100 GHz range for terrestrial services. (24 and 38 GHz are available now, while 28, 55 and 95 GHz are most likely next.). Systems in the near future will offer duplex service up to 622 Mbps, and high-speed (20 Mbps) Internet and data services. Beyond that, terrestrial FWA systems will allow data at 2 Mbps for third-generation systems, and even higher for future generations.

Terrestrial Mobile Wireless Access (MWA) cellular systems currently provide personal communication services (PCS) and low data rate digital services in the 800 MHz and 2 GHz bands.

For direct-to-user two-way mobile wireless services, future progress will be in the 0.8-6 GHz range. Some services in these bands are already available. The second and third generation mobile systems will be digital and will offer data rates as high as 384 kbps to pedestrians and 144 kbps to vehicles. They will have better quality of service (QoS) and may also enable position location of moving terminals.

Satellite Systems

Satellite based wireless systems also serve both fixed and mobile terminals. They do (or will) operate in L, S, C, Ku and Ka frequency bands (about 1.2, 3, 7, 12, and 18 GHz respectively) using satellites in geostationary earth orbit (GEO) as well as low or medium earth orbits. Although the Iridium system is no longer operational, others (such as Globalstar and Orbcomm using LEO, American and Australian Mobile satellite systems

using GEO) still exist, and several others (such as Skybridge, Teledesic, etc.) are planned. However, their continued operation or future implementation is not assured, because of the high initial costs and a limited customer base.

Satellite-based Fixed Wireless Access (FWA) systems, like terrestrial systems, currently operate the microwave and millimeter bands and offer duplex (simultaneous two-way) data rates of up to 155 Mbps, and simplex (one way) services such as television distribution.

For direct-to-user two-way mobile satellite wireless services, progress will continue in the 1-3 GHz range where services are already available. Future satellite systems, like terrestrial ones, will offer duplex service up to 622 Mbps, and high-speed (20 Mbps) Internet and data services.

Much satellite system research is also ongoing in areas of higher frequency bands (20-30 GHz and higher to 50 GHz), multiple high gain antenna beams, and transistors and traveling wave tubes with higher power and efficiency. New developments will include low-cost very small aperture terminals (VSATs) at higher frequencies, with interfaces to standard terrestrial networks. Large-scale satellite systems under development for such services include Teledesic, Spaceway and Astrolink; as an example, Teledesic will provide access rates of up to 2 Mbps to users and 150 Mbps to gateways. The systems will be able to handle both circuit-switched and packet switched traffic, including highly bursty Internet and multi-media traffic at high data rates.

Outlook for Wireless Systems

Satellites will have a 5 to 10 percent share of a national and global telecommunications infrastructure market of $180 billion and more than 113 million global users of broadband services in 2005.

The main technical challenges facing wireless systems are propagation-related signal degradations, such as: multipath effects; signal blocking by buildings and other obstructions; fading, attenuation, and cross-polarization due to rain; and effects of the atmosphere and ionosphere. Research is underway to compensate some of these effects.

Issues relating to spectrum allocation, standards, and interoperability will have a strong impact on the development and deployment of wireless systems. The International Telecommunication Union, several national regulatory agencies, and industry-based standards developing organizations are addressing these issues. Especially difficult are questions of obtaining additional spectrum for these services, and of frequency sharing (or band-segmentation if sharing is not feasible) for multiple systems providing multiple services in a given area.

Conclusions and Recommendations

Conclusions drawn from this background and policy recommendations flowing from the conclusions appear in the Introduction, above, and are not repeated here.

End Notes

(1)  See http://www.ieeeusa.org/COMMITTEES/CCP/harvard/index.html

(2)  Alan McAdams, Jean Camp, and Shastri Divakaruni. (2000, April). In Info 2(2):107-208, Special edition, The evolution of US telecommunications infrastructure.

The Institute of Electrical and Electronics Engineers, Inc.--United States of America
2001 L Street, N.W., Suite 700
Washington, DC 20036-5104
Phone: 202-785-0017, Fax: 202-785-0835.

| Top of Page | Position Statements | Policy Forum | IEEE-USA |

Last Update: 27 Feb. 2001
Staff Contact: Deborah Rudolph, d.rudolph@ieee.org