3/99 article. Settling New Territory. Ka-Band Update [Skybridge and Teledesic info]
satellitetoday.com
by Mark Williamson
IN 1995, A CLUTCH OF SATELLITE OPERATORS SOUGHT
APPROVAL TO OPERATE SO-CALLED BROADBAND SATELLITE
SYSTEMS USING A PART OF THE FREQUENCY SPECTRUM
KNOWN AS Ka-BAND. SINCE THEN, BROADBAND
MULTIMEDIA APPLICATIONS HAVE RECEIVED MUCH
PUBLICITY. BUT WHAT PROGRESS HAVE THE OPERATORS
MADE TOWARDS PROVIDING THOSE SERVICES?
When the wagon trains of the early American settlers trundled west toward a
land of promise and prosperity, their occupants' minds were focused on a
single goal: the acquisition of new territory, which they could own and profit
from in a way that had not been possible before. The satellite companies of
the 1990s that have recently embarked on a quest to develop a new tract of
the radio frequency spectrum known as Ka-band have much the same goal.
But developing a new and innovative satellite system is not easy at the best
of times, so why bother developing one that operates in the virgin territory of
a new frequency band? The answer lies in the term "broadband," sometimes
called "wideband," a reference to the width of the band of frequencies
available for communications services. With the lower allocations for C and
Ku-band already congested, many service providers are looking to the
frontiers and the unclaimed resource of Ka-band.
KA-BAND TECHNOLOGY
When a dozen or so satellite organizations filed for FCC approval in 1995 to
operate broadband satellite systems, the frequency of interest for some was
28 GHz, located in the upper part of Ka-band.
For convenience, Ka-band is often referred to as 30/20 GHz, where 20
GHz is the approximate downlink frequency and 30 GHz is the uplink, in the
same way that C-band is 6/4 GHz and Ku-band is 14/12 GHz, for example
(the higher portions of Ku-band are used for DBS services). The uplink is
almost always the higher of the two figures. Since it is more difficult to put
large antennas on satellites than in earth stations, the satellite is given the
easier job and allocated the lower frequency.
It is mainly the technical difficulties, and associated cost, of developing higher
frequency radio equipment that has constrained the industry to the lower
bands. The first communications satellites used C-band because radio
equipment was readily available at that frequency. Europe's first
communications satellites used Ku-band, partly in an attempt to leapfrog the
by-then outdated C-band technology.
Following this philosophy, the European Space Agency (ESA) began the
development of Ka-band systems in the late 1970s, culminating in the launch
of its Olympus technology demonstration satellite in 1989. Among other
things, Olympus carried a 20/30 GHz communications payload and a
Ka-band beacon payload to quantify rain attenuation at those frequencies.
Its successful operation proved the potential of Ka-band and led to the
development, by Italy, of the Italsat 1A and 1B satellites, designed to
demonstrate the operational capabilities of an advanced Ka-band payload
and provide a pre-operational service within the Italian telecommunications
network. The Italsat spacecraft were launched in 1991 and 1996
respectively, while research into Ka-band was further progressed by
NASA's Advanced Communications Technology Satellite (ACTS),
launched in 1993, and Japan's Communications Engineering Test Satellite
(COMETS), launched in 1997.
KA-BAND APPLICANTS
The first companies to apply to the FCC for Ka-band capacity for
operational broadband systems were Hughes Communications Galaxy Inc.
(for its Spaceway system); Kastar Communications Corp. (for a single
satellite); Loral Aerospace Holding Inc. (for its Cyberstar system); Panamsat
(for PAS 9); and Teledesic Corp. (for its then 840-satellite, now
288-satellite, constellation).
A number of others subsequently jumped aboard the bandwagon, ensuring
they met the FCC's application deadline. They included GE Americom with
GE*Star, Lockheed Martin Corp with Astrolink, AT&T Corp. with
Voicespan, Motorola Corp. and Vebacom GmbH of Germany with
Millennium and a number of single-satellite applicants. Since then, Voicespan
and some of the smaller systems have been disbanded, while Millennium,
and another Motorola proposal called Mstar, have been combined to form
Celestri, which, in turn, was merged with its erstwhile competitor, Teledesic,
in 1998.
It was always expected that many of the original applicants would fall by the
wayside, but the departures have left a significant number to forge ahead
with their technical, organizational and financing programs. The applicant that
started the bandwagon rolling was Teledesic LLC, famously backed by
Microsoft chairman and CEO Bill Gates and telecommunications pioneer
Craig McCaw, who is also Teledesic's chairman.
Teledesic is developing what the company calls a global, broadband
"Internet-in-the-sky." Using a constellation of 288 low earth orbiting
satellites, it will be the first satellite communications network to enable
"affordable, worldwide access to 'fiber-like' telecommunications services
such as broadband Internet access, videoconferencing and interactive
multimedia," according to the company.
Teledesic's Steve Hooper, who has shared the title of co-chief executive
officer with Craig McCaw since December 1997, is upbeat on the system's
status. "We're making great progress behind the scenes on our system and
partnering efforts," he says. "We've been working closely with our prime
contractor, Motorola, to finalize all aspects of our system and formalize the
roles of all our major subcontractors and other partners."
According to Hooper, quality of service will be the "fundamental
characteristic" that differentiates broadband satellite networks from the
others, just as it is with terrestrial broadband networks. "Teledesic's ability to
provide guaranteed end-to-end quality of service anywhere in the world will
continue to set us apart," he says. "Our ability to execute our plans in 1999
will bring us closer to realizing the world's first Internet-in-the-sky."
Another leading contender for the broadband market is Hughes
Communications' Spaceway system. Having made an initial application for a
15-satellite system, the company announced, in December 1997, an
enhancement to increase overall system capacity and add higher data rate
transport services. Its dual FCC filing covered Spaceway EXP for an
eight-satellite system operating in geostationary orbit, and Spaceway NGSO
for a 20-satellite system in non-geostationary orbit. Five NGSO satellites
would orbit in each of four planes inclined at 55 degrees to the equator at an
orbital altitude of 10,352 km.
According to a company statement, both systems are designed to operate in
the Ka-band frequency range (17.7 GHz to 30.0 GHz). Spaceway EXP will
focus on the high data rate transport market using satellites in four orbital
locations (117øW, 69øW, 26.2øW, and 99øE were requested), while
Spaceway NGSO will add broadband capacity at a wide range of data
rates.
The satellites themselves are intended to incorporate multiple-beam
antennas, digital processors for switching traffic among beams and
intersatellite links (ISLs) to interconnect satellites. This means that a signal
received by one satellite can be relayed directly back to the same beam,
switched to another beam or relayed by ISLs to other satellites.
At the time of the announcement, Ed Fitzpatrick, HCI vice president for
Spaceway, stated that the company hoped "to move out as planned with the
original Spaceway architecture as licensed by the FCC in May 1997." Once
licenses for the two new filings were received, he added, "we expect to use
them to expand our service offerings as the market demands."
As proposed, the global Spaceway system would be able to provide
coverage in four main regions: North America, Asia Pacific, Latin America
and Europe, Africa and the Middle East. Its "bandwidth-on-demand"
capability would provide businesses and consumers with fast access to
terrestrial networks, such as the Internet, Intranets and local area networks
(LANs). The system would use a family of receive/transmit antennas as small
as 66 cm in diameter and provide uplink speeds of up to 6 Mbps.
The company puts data rates in perspective by comparing the capabilities of
a 384 kbps Spaceway link with a conventional telephone line. Whereas a 1
Mbit digitized photograph can be transmitted down a phone line in about 34
seconds, the satellite link would transmit it in just 0.7 seconds. Comparable
times for The Washington Post Sunday edition would be nine minutes and
10.4 seconds, respectively-figures that speak for themselves.
As of January 1999, Hughes was refining a forthcoming statement regarding
the Spaceway system, but confirmed that the company had "removed all
doubt regarding its commitment to the broadband, Ka-band market" by
announcing that its board of directors had approved a $1.35-billion-dollar
investment to create the business in North America. As a result, the
company expects to launch two satellites, which will allow it to begin service
in North America in 2002.
Interestingly, Hughes entered the Ka-band arena in October 1998 with the
launch of the U.S. Navy's UHF Follow-On satellite, UHF 9, which
incorporated a Global Broadcast Service (GBS) operating at Ka-band.
According to Hughes, the satellites are among the first operational systems,
government or commercial, to carry Ka-band payloads.
According to Spaceway, its North American business development is "a
synergy among all Hughes units, with Hughes Space and Communications
Co. building the satellites, Panamsat providing satellite operations, and
Hughes Network Systems and DirecTV ultimately providing key end-user
marketing and distribution outlets."
From Panamsat's point of view, considering the ever-increasing need for
extra bandwidth, Ka-band is an important resource. The company's chief
technology officer, Robert A. Bednarek, says although the digitization of
content has made it easier to transmit information, "we still need the space
for it, [so] the development of Ka-band spectrum is very important." The
additional spectrum will aid in the further advancement of satellite service
offerings, he says, while the higher frequencies "theoretically allow for smaller
antennas on the ground and can lead to a more ubiquitous service."
Bednarek believes that it is important to remember, "Ka-band is not a
product but part of a spectrum." As with C-band and Ku-band, "which
support multiple applications like video, business communications and
Internet access, there are a variety of uses for Ka-band," he says.
According to Bednarek, while Panamsat currently has no satellites with
Ka-band payloads, in orbit or under construction, it is "actively designing
systems and satellites for that frequency." The company is also developing
specific Ka-band applications, including broadcast services and "special
telecom services," he adds, as well as exploring the option of using both
GEO and non-GEO satellites.
Another Ka-band applicant is Astrolink, a strategic venture of Lockheed
Martin, designed ultimately to operate using nine satellites in five orbital
locations to provide "a broad array of digital communications," including
voice, data and video. In fact, according to Astrolink, only four satellites are
required to provide global multimedia services, the first of which is planned
for launch in 2001, with initial operational capability expected in early 2002.
Astrolink's president and CEO, Celso Azevedo, is optimistic about his
company's prospects: "1998 has shown a substantially increased awareness
of the growth opportunity in global broadband multimedia services sparked
by Internet and e-commerce," he says. "In 1999, we will continue to see a
consolidation of the industry and continued mergers and partnerships, which
will reduce the number of satellite ventures seeking investment financing. This
should make financing easier for satellite systems, such as Astrolink, which
have a clearly defined business and technical strategy and strong technical,
operational and financial backing from their partners."
According to Azevedo, Astrolink will be announcing its international partners
and service providers in 1999, "clearly becoming the first venture to offer
global, wireless broadband telecommunications services to business
customers." As such, he adds, "Astrolink will be the catalyst that defines the
global Ka-band broadband satellite service industry."
NEW ENTRANTS
There will, however, be plenty of competition. Kastar Satellite
Communications Corp., a Denver, CO-based company and a relatively new
entrant to satellite communications market, is one of them. According to a
company statement, Kastar plans to build, launch and operate a global,
digital, Ka-band satellite system that will provide on-demand, low-cost,
two-way interactive multimedia services to operators of cable modem
systems, DBS satellite systems and other terrestrial network systems, initially
across the United States and subsequently around the world.
The company, which was incorporated in 1995, says the Kastar satellites
are unique because they provide extremely flexible broadband capacity at
low cost, using spotbeams and onboard processing. "No previous satellite
systems have had this flexibility or capability," says Kastar. The proposed
satellites are destined for the 109.2øW, 73.0øW, 52øE and 175øW orbital
positions.
What the company calls "just-in-time bandwidth" will provide international
connectivity to the Internet and other private and public networks at data
rates from 64 kbps to 155 Mbps. "As people learn to use and economically
benefit from higher data rates," says CEO and President Thomas E. Moore,
"Kastar expects higher data-rate connectivity from 1.5 to 5 Mbps."
Kastar filed its initial FCC application in July 1995, and an amendment the
following September, to construct, launch and operate a Ka-band digital
domestic fixed communications satellite system. The FCC assigned the
company its first two orbital positions-73øW and 109.2øW-on May 8,
1997; the 52øW and 175øW positions are pending. According to Moore,
Kastar intends to launch its first two satellites "prior to the year 2003."
In September 1998, Space Systems/Loral announced that it signed an
agreement with Kastar Satellite Communications Corp. to construct two
advanced Ka-band spotbeam satellites. The first satellite, in a contract
valued at more than $300 million, will be delivered in orbit no later than
February 2002, according to the company.
Another of the early applicants for Ka-band frequencies was Loral's
Cyberstar, which received its FCC license in May 1997. In June 1997,
however, Loral Space and Communications and Alcatel-Alsthom
announced a "cross investment" in their respective broadband
communications systems, Cyberstar and Skybridge. Although the companies
will remain separate, it is intended that they will pursue joint marketing
activities. Previously, Cyberstar planned to use three Ka-band GEO
satellites; Alcatel's $3.5 billion Skybridge, previously named Sativod, will
use a constellation of 80 Ku-band satellites in low earth orbit.
According to Pascale Sourisse, president and chief executive officer of
Skybridge Limited Partnership, use of the well-established Ku-band reduces
technology risk, the price of system components and the potential for signal
fade. Moreover, locating the satellites in LEO produces lower signal
propagation delay than GEO and will allow the use of smaller end-user
antennas.
Skybridge will use a unusual spectrum-sharing scheme, developed by the
company to solve potential interference problems, which involves switching
off the transmissions from the satellites in LEO as they cross the
geostationary arc. The handover procedure is ensured by the gateway and is
fully transparent to the user. Other satellites in the constellation would take
up the signal at this point.
Skybridge filed its space segment application with the ITU in 1995/96 and
with the FCC in February 1997, receiving approval from the ITU at the
World Radio Conference (WRC) in November 1997, and what the
company considered "a positive sign" from the FCC in November 1998.
Pascale Sourisse expressed her pleasure at the decision which, she says,
represented "another milestone for Skybridge, as we advance toward our
goal of commencing service in 2001."
An industrial team under the leadership of Alcatel is now in place to design
and develop Skybridge, with more than 400 engineers working on the
program. "Skybridge is right on track with its development strategy and
schedule," says Sourisse.
BROAD INTEREST
Despite Skybridge's decision to stick with Ku-band, the large number of
companies that have shown interest in developing Ka-band systems,
satellites or payloads is indicative of the future importance of the band.
For example, in May 1998, GE Americom entered into a contract with
Harris Corp. for one Ka-band satellite with options for additional
spacecraft, much to the surprise of many observers because it marked the
emergence of Harris as a satellite prime contractor. Since then, according to
Monica Morgan, a spokeswoman for GE Americom, the two companies
have been engaged in trade-off and optimization studies to determine the
best payload design for the range of markets that are being identified in
parallel market research efforts.
Before either of these efforts can be concluded, however, GE must await the
results of an FCC edict, which proposes to reallocate half of the frequency
bands. According to Morgan, the proposed spectrum reallocation is "not an
adequate resolution of Ka-band spectrum for Americom," and the issues
"must be resolved before final definition of the system architecture can be
completed."
Soci‚t‚ Europ‚ene des Satellites (SES), the operator of the Astra system, is
also pursuing Ka-band developments, as shown by the fact that Astra 1H
and Astra 1K will carry two Ka-band transponders each in addition to the
usual stack of Ku-band transponders. In December 1998, SES signed a
contract with Nortel Networks for the provision of a turnkey interactive
satellite system, which, according to SES spokesman Yves Feltes, will be
"Europe's first commercial Ka-band satellite return channel system."
Currently, SES's Astra-Net multimedia platform uses digital video broadcast
(DVB) technology at Ku-band to provide customers with multimedia data at
rates up to 38 Mbps. From 2000, the Ka-band transponders on Astra 1H,
and from 2001 Astra 1K, will provide the return path at the higher
frequencies (29.5-30 GHz/18.3-18.8 GHz), allowing interactive broadband
and bandwidth-on-demand services.
SES's arch rival, Eutelsat, is also interested in Ka-band, as indicated by the
publication of at least 20 Ka-band orbital position requests with the ITU
back in 1996.
The following year saw the announcement of another two European
Ka-band proposals. One, Euroskyway, is backed by Alenia Aerospazio
and two dozen partners, and will eventually comprise five dedicated
satellites, the first of which should be launched in 2000. The other, an
initiative for an advanced Ka-band digital satellite from Matra Marconi
Space called WEST (Wideband European Satellite Telecommunications), is
intended to be used for interactive wideband multimedia applications, which
could include interactive TV, telemedicine, telebanking and investment
services, and even interactive gaming and gambling. If partners, and more
importantly funding, can be found, the $2 billion WEST system could
eventually comprise at least one GEO satellite covering Europe and up to
nine ICO (intermediate circular orbit) satellites to provide global coverage.
Although they have yet to progress beyond the proposal stage, these
projects could, if successful, herald a new era in satellite contracting; one in
which manufacturers offer the market their most innovative technology,
rather than responding to requests for proposals from prospective
customers.
HIGHER, BETTER?
Although Ka-band is more susceptible to rain attentuation, when compared
with the increasingly crowded lower frequency bands, Ka-band begins to
look like California did to the settlers from the East, offering broad tracts of
bandwidth to accommodate the spectrum-hungry applications of the new,
multimedia/Internet age of telecommunications.
But what is that, on the misty horizon? In late 1997, another stack of
frequency applications was received by the FCC-this time for V-band
frequencies (40 to 50 GHz). Licensing procedures have still to be
determined, but the activity suggests that yet another round of financing,
contracting and posturing is about to begin. It also shows there is rarely a
pause in the development of satellite communications; like the frequency
spectrum itself, it is a continuum with no discernible end.
Mark Williamson is an independent space technology consultant to space
industry and insurance sectors and a technical author. He is based in the
United Kingdom and can be contacted at +44/1768-361-040 (tel/fax) or at
markwilliamson1@compuserve.com.
Back to March 1999 Issue
Copyright © 1999 Phillips Publishing International, Inc. |
