Achieving the Mobile Internet

March 2001
By Ken Keane

Experts debate accommodation of third-generation communications devices.

Discussions between commercial and government organizations about spectrum management are coming to a head as third-generation wireless devices that combine today’s desktop terminal features and functions in a shirt-pocket-sized handset are poised to enter the marketplace. This communications evolution poses some critical issues for regulators as well as the military. Experts not only must confront the traditional spectrum allocation challenges, but they also must consider separation, partitioning, sharing and reallocation. The policy issue may reduce itself, on the one hand, to minimizing the costs of securing new spectrum for third-generation devices and, on the other hand, to the desire for greater global harmonization.

While a number of options are being explored, spectrum for so-called third-generation wireless telephones, also known as 3G or IMT-2000, could be allocated from existing resources. The federal government has already transferred some spectrum to the private sector, and an additional source also exists. In essence, the current supply could be pieced together to meet the growing demand.

First-generation (1G) wireless devices were the analog cellular telephones fielded in the United States in the early 1980s. These devices operate in the 800-megahertz range. Compared with today’s digital telephones, 1G cellular telephones lacked features such as voice messaging, were bulky and were expensive, initially costing as much as $3,000.

Much of the equipment in use today is second generation (2G). These digital devices, first fielded in Europe and Asia, operate primarily in the 1.8- to 1.9-gigahertz range. They are small, include a multitude of features and are often provided free of charge as part of a service contract.

While 3G devices also will be digital, in important respects the comparison with earlier generations ends there. As planned, 3G devices will combine many of the functions of today’s computers and put them in the palm of the user’s hand. In effect, the Internet will go mobile.

Proponents of 3G devices point to the fact that the number of 1G and 2G cellular users in the United States has grown from virtually nothing to 100 million in less than 20 years, with growth continuing at a rate of 25 to 30 percent annually. Internationally, there are now some 470 million cellular telephone users, a number that is expected to increase to 1.3 billion in the next five years.

At the same time, Internet usage has skyrocketed. Ten years ago, use of the Internet was largely confined to academia and the defense establishment. Today, more than 50 percent of U.S. households have Internet access, and Internet usage is doubling every 100 days.

The potential of 3G has not been lost on the international community. The International Telecommunications Union (ITU) estimates that in 10 years there will be more mobile than fixed telephones worldwide. Faced with these projections, the ITU has called upon nations worldwide to study the possibility of additional spectrum allocations to accommodate 3G.

At the 2000 World Radiocommunication Conference held in Istanbul in May 2000, the ITU adopted a resolution that focuses on possible global allocations from three bands: 1710 to 1885 megahertz, 2500 to 2690 megahertz, and 806 to 960 megahertz. However, the ITU also recognized that the 698- to 806-megahertz band could be considered for 3G services and agreed that each nation has full flexibility to decide whether and how to allocate additional spectrum for 3G.

The spectrum from 1755 to 1850 megahertz is a government band utilized for a variety of U.S. national security purposes. The most prominent of these is the satellite ground link subsystem (SGLS). The SGLS network controls scores of military satellites used in communications, intelligence gathering, nuclear test detection, missile launch warning and other missions. SGLS also controls the global positioning system. In addition, this band is used for air combat preparation, including final readiness training before deployment to combat in high-risk areas, U.S. Army and U.S. Marine Corps tactical radio systems, precision guided munitions and a variety of point-to-point microwave links. The federal government has characterized the 1755- to 1850-megahertz band as critical for U.S. military operations, particularly because of its favorable propagation characteristics.

Schools and colleges use the 2500- to 2690-megahertz band for distance education in the delivery of instructional programming. As the result of a 1998 policy change by the Federal Communications Commission (FCC), these systems also may be used for the delivery of two-way, high-speed Internet access services. The potential value of the spectrum in a two-way configuration has not gone unnoticed. Sprint and WorldCom have invested more than $2 billion in acquiring the rights to use excess capacity that is not being utilized by educational licensees. In furtherance of this policy change, the FCC opened a filing window in August 2000 for two-way applications. Scores of applications were filed in that window looking toward the implementation of two-way Internet delivery using the educational spectrum.

The U.S. Commerce Department’s National Telecommunications and Information Administration (NTIA) and the FCC have been conducting parallel studies of 3G allocation, each focusing on the band within its respective jurisdiction. The NTIA is examining the 1755- to 1850-megahertz band, and the FCC is studying the 2500- to 2690-megahertz band.

Meanwhile, the White House also has weighed in on the issue. On October 13, 2000, President Bill Clinton issued a memorandum directing all executive branch agencies to cooperate in the search for 3G spectrum and setting a timetable for action. In addition, he strongly encouraged the FCC to complete its studies. Issuance of the president’s memorandum is an extraordinary development. Observers recall no other instance when a president has plunged into the middle of a spectrum allocation matter more directly and more conspicuously than in this case.

On November 15, 2000, the FCC and the NTIA simultaneously released interim reports on the study of their respective bands. The FCC’s report concludes that “large co-channel separation distances are needed between 3G systems and [educational] systems to avoid causing harmful interference to [those] systems.” It recommended a distance on the order of 100 miles. The report goes on to state that the sharing prospects between educators and 3G are “extremely problematic”; that “voluntary partitioning between users and 3G operators…could offer some promise of sharing as an interim measure”; and that, while “options for increasing the spectrum efficiency of [school] operations must be explored,” any such improvements probably would not be enough to offset the loss of spectrum from a 3G reallocation.

The NTIA’s report likewise is guarded about the chances for 3G sharing. It indicates that “the most serious challenges in accommodating [3G] systems are related to sharing with federal government uplink satellite control systems, the military radio relay systems and the air combat training systems.” In addition, it notes that “uplink satellite signal margins would be severely degraded by the aggregate [3G] transmissions” and that co-channel sharing between 3G and SGLS, air combat training systems and mobile subscriber equipment would require large separation distances. The NTIA report then considers various options under which the band could be segmented, that is surrendered in part, to accommodate 3G operations.

The NTIA’s conclusions are based heavily on U.S. Defense Department studies of the prospects for sharing spectrum with 3G rather than the NTIA’s own analysis. The effect, if not necessarily the intent, is to place some distance between the NTIA and the sharing conclusions contained in its report. This illustrates a tension inherent in the NTIA’s role. On one hand, it is the spectrum policy maker for the administration; on the other hand, it is the spectrum advocate for executive branch agencies such as the Defense Department. Given the major White House involvement in the 3G matter and the political interests involved in the debate, it is not surprising that the NTIA is proceeding cautiously.

Meanwhile, the FCC has released a notice of proposed rule making that offers certain suggestions of its own for 3G allocations. Among other things, the commission has expressed the view that current cellular bands could evolve to 3G from the 1G/2G services offered today. Those bands include 806 to 960 megahertz and 1850 to 1990 megahertz. The notice asks for comment on the possibility of allocations from the 1.7- and 2.5-gigahertz bands, and it opines that 30 megahertz from the 700-megahertz band—spectrum already slated for auction this spring—would also be available for 3G.

The next step in the reallocation process is for the NTIA and the FCC to examine the costs of relocating incumbent users. These include costs that would be imposed on government licensees displaced in the case of the 1755- to 1850-megahertz band and costs that would be imposed on schools and colleges displaced in the case of the 2500- to 2690-megahertz band. In addition, the NTIA and the FCC are to report on alternative bands to which their respective user groups could be relocated. These reports were due out by March 1, with a final allocation decision by July 2001, and 3G licenses auctioned by September 30, 2002.

Under the law and regulatory policy, incumbent licensees evicted as part of a spectrum reallocation may expect reimbursement for the costs necessary to provide comparable communications capacity in alternative bands. In the case of the 1755- to 1850-megahertz band, reimbursement is a matter of law. Specifically, the Strom Thurmond National Defense Authorization Act of 1999 requires private sector users bumping government incumbents to pay the costs of relocating the government users. Thus, the law ensures that the taxpayer is not called on to subsidize the new entry by private parties. The NTIA is currently conducting a rule making that looks toward implementation of this law.

In addition to the reimbursement issue, another provision of the law stands as a potential stumbling block to reallocation of the 1755- to 1850-megahertz band. Section 1062 of the National Defense Authorization Act of 2000 requires that, before any Defense Department spectrum user can be evicted, the NTIA must make alternative spectrum available. Moreover, the chairman of the Joint Chiefs of Staff and secretaries of defense and commerce must certify that the prepared alternative spectrum will provide “comparable technical characteristics to restore essential military capability.” Section 1062 shapes up as a central issue in any further examination of Defense Department spectrum.

While there are a number of facets to this issue, one of the more unusual thus far is the failure to actively consider the use of the 698- to 746-megahertz band for 3G. This band comprises current analog television channels 52 to 59 and is scheduled for auction by September 30, 2002, incident to the conversion of television broadcasting to digital modulation. In effect, the FCC and the NTIA decided to tackle the hard issues of the 1755- to 1850-megahertz and 2500- to 2690-megahertz bands, while discounting a simpler solution.

The ITU has about 160 additional megahertz that may be needed for 3G by the year 2010. Eighty-five megahertz of that requirement is already available, specifically the 1710- to 1755-megahertz and the 2110- to 2150-megahertz bands. Taking note of this, the FCC has proposed allocating these bands, along with 2160 to 2165 megahertz, for 3G. This produces a combined total of 90 megahertz. The remaining spectrum needed to total 160 is available from the 698- to 806-megahertz band. In particular, 78 megahertz would be available with the 30 megahertz targeted for auction this spring and an additional 48 megahertz from the adjacent 698- to 746-megahertz band. Added to the 90 megahertz, policy makers would have secured more than 160 megahertz.

The television spectrum will not be available for nonbroadcast purposes until the end of 2006. In fact, current law provides that television broadcasters need not vacate the spectrum even then unless digital set penetration has reached 85 percent. In addition, the band may not offer as much global harmonization as other bands. However, FCC officials have already begun exploring the possibility of changing the law to deal more effectively with the reluctance of television broadcasters to relocate—the so-called “spectrum squatter” problem. Moreover, the spectrum may not be needed for 3G until the latter half of the decade, by which time substantial relocation by television broadcasters is likely to be an accomplished fact. Finally, the ITU has recognized 698 to 806 megahertz as one of the bands suitable for 3G. In addition, the FCC itself has said “it appears very unlikely that a single band plan can be adopted on a global basis.”

The competition for spectrum between the military/academia on the one hand and commercial interests on the other hand is likely to persist. As the technology matures and 3G devices become more readily available, government decision makers will continue to be faced with balancing the requirements of the military and educational communities with the demands of commercial interests.


Ken Keane is an attorney at Arter & Hadden LLP, Washington, D.C., who specializes in the practice of telecommunications law and policy.