The Big Five: Celestri, CyberStar, SkyBridge, Spaceway and Teledesic
(Profiles are drawn from interviews with company executives or partner company executives; for more information on these companies see chart.)
Loral's Different Path: Buying, Phasing, Emphasis on Business
If any satellite provider is taking a road less traveled, it's Loral Space & Communications Ltd. Under chief executive Bernard Schwartz, Loral is primarily buying, not building, its way into the broadband
satellite arena. It is also putting its focus on business and consumers, rather than dealing exclusively with carriers and ISPs.
Loral plans to phase in its CyberStar services through leased facilities over the Skynet satellites it purchased from AT&T for $478.1 million. Testing of this system began last fall and should be ready for full service broadcast at 30Mbps by summer. The network operations center now being tested will determine-based on pricing and type of application-whether to direct traffic to the Internet or satellite services.
A second phase-which seems more and more tentative-calls for providing two-way Ka-band service at 500Kbps uplink and 3-6Mbps downlink. Timing for that phase is now hovering at the year 2000 or beyond, if at all.
As CyberStar brings on services Loral's celestial ownership and partnership presence will continue to expand. Today, Skynet includes three satellites, but Loral will be replacing one of those aging satellites this year and plans to add two
more satellites to the constellation at the beginning and end of 1999.
At least three more Ku-band GEOs will arrive from the finalization of Loral's purchase of Orion, expected early this year. One of those GEOs is already in orbit, a second is slated for this year and a third for 1999. CyberStar President Ron Maehl says CyberStar will tap Orion's groundstation facilities to extend its services to Europe and the Middle East.
Loral's expanding international presence also comes in the form of 75 percent stake with Telefonica Autrey in Satellites Mexicanos, S.A. de C.C. and its 39 percent stake in Globalstar, a 56-satellite mobile voice system that will compete with Iridium and ICO.
Through a $30 million cross investment with Alcatel Alsthom's SkyBridge 64-satellite LEO constellation, Loral plans to pursue integrated marketing to better serve real-time and interactive applications. Finally, Maehl says Loral plans to build its own GEO constellation of three satellites and one spare (estimated
before the FCC at $1.05 billion). But the goal, here, is to "develop the market" before deploying these added birds, he says.
The bottom line on these many satellite investments is that Loral will eventually be positioned to offer some tightly woven, multi-service marketing packages.
Maehl expects two-way PC terminal conversions and equipment for CyberStar to cost about $1,300, with the cost cut to about $300 for receive-only units. For large businesses the company is exploring using $150 PCI cards to communicate with a central receive dish. Server-based communications are also being considered.
To Sell or Resell
One of the truly unique aspects of CyberStar's business plan is its emphasis on businesses and consumers in addition to selling to other service providers. In fact, Maehl says the businesses most interested in CyberStar's upcoming services tend to be large Silicon Valley companies, server and network systems providers, and the more traditional value added network providers a
nd carriers. The service is focused on the Americas, Europe and Asia.
Loral's purchase of Orion plays particularly well into CyberStar's end-user emphasis, since Orion already serves about 260 private businesses and ISPs in 47 countries.
The end-user emphasis may also lie behind Maehl's belief that security is of "primary importance" to CyberStar. He says the service will include smart-card based authentication, although a decision has yet to be made on encryption.
Most broadband satellite providers, like Celestri and Teledesic, are either taking a strictly wholesale outlook or saying that they're willing to accommodate a few extremely large businesses or organizations. In many instances their logic is that it can be difficult to convince national providers to give up frequencies and allow satellite competition on their own turf unless they give national providers a piece of the action. So, most satellite providers say they'll partner with existing national providers-even when they have
their own intersatellite links that technically permit bypass of national entities.
Maehl says CyberStar is still examining the possibility of ISLs with SkyBridge and its own Telstar satellites, but leaning away from the technology because of technical problems associated with multiple spaceborne hops and the fact that ISLs use up critical power that could be directed toward the earth.
Finally, Loral has an edge with its affiliation with Space Systems Loral, which is able to buy launch vehicles in bulk.
Teal Group analyst Marco Caceres says that "Loral is committed to being a space power," adding that he wouldn't be surprised to see Loral purchase even more satellite systems as time goes day.
Loral is also looking beyond Ka-band to higher frequencies in the more distant future. Recently, Loral applied to the FCC for higher frequencies for its CyberPath, a $1.2 billion system of 10 GEOs (four were included in the application) that would rely on $1500 earth stations for broadband com
munications.
Motorola Seen As One of the Safest Bets in Space
If any company will prevail in the face of an impending broadband satellite bloodbath it's Motorola's Celestri. That's the consensus among most satellite watchers. The reasons are many-Motorola's financial standing, its experience in the industry, its reputation for quality, the multinational negotiation experience it gained as a lead player on the mobile voice Iridium constellation, and its decision to build a system that takes advantages of the economies of GEOs coupled with the anticipated quality of LEO transmission.
Bary Bertiger, corporate vice president and general manager of Motorola's Satellite Communications Group, expects Celestri to appeal to small and mid-sized business "because it gives them the global connectivity of a large business." He also expects large multinationals that need to reconfigure their sites often or that need to bring up sites quickly to turn to Celestri.
Bertiger says Motorola has looke
d at companies, including its own, and found that there isn't a "a clearly compelling" argument that would cause these companies to give up existing leased line services for the cost savings they'd gain in going to satellite. However, he says as the services moves into more and more places-especially China-"the economics get more interesting." He expects broadband satellite to be most competitive for new services.
Motorola's consolidated terminals are expected to cost about $750 and look a lot like a miniature weber grill we're told.
Motorola's Celestri incorporates Motorola's previously planned M-Star
and Millennium constellations, with Celestri satellites carrying a dual
Ka and V-band payload. The V-band payload is expected to be used
primarily for high-bandwidth applications such as trunking.
-------------------------------------------------------------------------------------
Hughes Goes Global With MEO Expansion to Spaceway
Hughes moved up with the K
a big boys recently when it announced that it was expanding its eight GEO Spaceway to become a 16 GEO, 20 MEO global constellation.
The Hughes Communications Inc. expansion is apt to zip up analysis suggesting that Hughes-one of the world's largest satellite players today-will be a secondary player in second generation broadband systems.
By 2001, Hughes expects to launch one to four GEOs from an eight-GEO base costing about $3 billion. Eventually the constellation is expected to total 16 GEOs and 20 MEOs. And those last eight GEOs, are expected to pack a punch-60Gbps duplex per satellite versus 35.2Gbps collectively for the first eight. These later GEOs are part of the EXP expansion, while the MEOs come under what is called the NGSO expansion. Both EXP and NGSO are Ka-band systems. Hughes will rely on onboard processing, spotbeam technology and intersatellite links in its expansions.
Hughes decided to invest in global MEOs in an inclined orbit of about 10,352 kilometers because of "thei
r economics." The MEO expansion is targeted at latency-sensitive applications and calls for antennas of about 12.5x12.5 inches for up to 2Mbps; about 20.5 inches for up to 10Mbps; and 2 meters for up to 155Mbps. The GEO expansion is primarily aimed at intercontinental and intracontinental trunking and multi-casting at up to 155Mbps, with 99.99 percent availability and 3.5 meter terminals.
Hughes VP Edward Fitzpatrick says that the advantage of Spaceway over existing systems like DirecPC is improved connectivity, improved bandwidth and data rate, greater versatility and smaller antennas. For example, he says businesses will be able to tap up to 6Mbps from Spaceway using antennas that are only about 26 inches wide. Those terminals are expected to cost about $1,000 in volume, but Fitzpatrick says that the terminal pricing won't be allowed to "get in the way of driving volume on the ground."
Fitzpatrick primarily envisions the base system being used for remote access to ISPs or LANs for one to three mi
llion subscribers. When it comes to competing with T1, he expects the system to be most competitive with users who need six or less T1 duty hours per day. He sees Spaceway as being similar to xDSL, but believes that the xDSL rollout will be spotty and in heavily populated areas where it will make the most money. The advantage of Spaceway, he says, is its ubiquity. Fitzpatrick expects the biggest competition on the business side to come primarily from xDSL and wireless technologies.
Hughes has also applied to the FCC for v-band frequency to be used in a follow-on higher-speed system known as Expressway. Fitzpatrick says to look for Expressway in the 2004 to 2005 timeframe. Hughes expects Expressway to be used primarily for high-speed point-to-point trunking.
Finally, Hughes owns a majority share in PanAmSat Corp., is one of the largest existing satellite providers. PanAmSat actively promotes ISP caching over its constellation. For example, one of Japan's largest ISPs is working with PanAmSat aggr
egating Internet traffic in the U.S. and delivering it via antennas to smaller ISPs. PanAmSat has 17 satellites and plans to launch six more by late 1999. The company has also requested Ka-band frequency for a GEO satellite. PanAmSat, however, declined to participate in our satellite survey.
SkyBridge: Something Old, Something New, in the Sky so Blue
Alcatel's SkyBridge may not be at the top of most U.S. analysts' best picks lists, but the $3.5 billion constellation is interesting in its own right for blending old and new technology.
While SkyBridge is new in the sense that it's a 64-bird LEO constellation without the delays inherent in GEO systems; it's old in the sense that it intends to rely on Ku-band frequencies. SkyBridge also plans to use rely extensively on traditional ground-station communications versus intersatellite links.
Whether this blend proves a winning strategy remains to be seen. Certainly, system complexity is reduced without ISLs and the political job o
f securing access rights around the globe becomes easier if existing providers tie into some 200 ground stations planned for SkyBridge. SkyBridge, too, is counting on building from a base of existing ground terminal equipment that has already been tried and found to be true. Finally, it won't have to deal with the power requirements of new Ka birds.
The downside to this approach is that SkyBridge (and other constellations like Celestri that may share spectrum) must deal with the complexity of having two birds in sight whenever interference with existing frequency allocations becomes a possibility. Analysts also suggest that spotbeam reuse of frequency is more difficult in the Ku band and may prove limiting from a bandwidth perspective (Each satellite will illuminate a 300 kilometer radius area with up to 45 spotbeams illuminating a 350 kilometer radius). Some analysts also point to SkyBridge's consortium lead, Alcatel, as having less of a name in space than players like Hughes.
Yet, SkyBridge al
so has its fans. Leslie Taylor, president of Leslie Taylor Associates, ranks Celestri and SkyBridge as her top two broadband satellite picks. She believes there's something to be said for simplicity of design and easy transition like that promised by SkyBridge. Alcatel's ATM know-how and its big league presence in Europe (although SkyBridge is technically incorporated in Delaware) shouldn't hurt either. And SkyBridge has rounded up a large number of financial backers, including important Asian companies like Toshiba. Finally, SkyBridge has a $30 million mutual cross-investment with Loral that intertwines the companies' marketing approaches.
Service is scheduled to begin toward the end of 2001(with half the planned constellation) and aggregate capacity is set at 144Gbps (up to 2Mbps uplink and 60Mbps downlink). Terminals will range from personal units to those designed for residential or corporate buildings and include an outdoor component as well as a system interface such as a PC, set top box, PABX
or other device.
A SkyBridge spokesman says the constellation has a conservative target of 20 million users. Services are targeted to North America, Europe and parts of Asia initially. The spokesman says most of SkyBridge's partners don't expect to have true broadband terrestrial networks before 2008 or 2010 because of the huge investment and time a terrestrial offering will take. He says they see SkyBridge as a way to bring such services to customers in the interim. In January, the ITU had already approved SkyBridge frequencies and the consortium was awaiting an FCC license.
---------------------------------------------------------------------------------------------------------
Debating the Stars: Teledesic is in the Thick of It
Teledesic is perhaps the best known-and most debated-of the next generation satellite providers. Much of its visibility stems from its well-heeled investment duo, Craig McCaw and Bill Gates; its respected former architect and present consultant Hans-Werner
Braun; and Boeing's investment and integration role for the venture.
Teledesic generates some of the most vociferous admirers and detractors of any of the next generation satellite players-probably because of the breath of the effort, Teledesic's willingness to up the ante in its lobbying tactics.
However, Boeing's role in cutting the Teledesic constellation from an unwieldy 840 to 288 birds has muted a number of Teledesic's critics, especially since many expect to see the project downsized even further. It's certainly the largest LEO constellation proposed, but whether it's too ambitious still remains to be seen. From a technical standpoint, Teledesic's size should prove beneficial in minimizing variable latency concerns, and certainly Teledesic's low earth orbit is aimed at providing quality superior to GEOs for interactive applications.
The big questions seems to lie with its newcomer status to the industry, the scope of its effort and concerns that new financial partners don't seem t
o be visibly buying into the project beyond Craig McCaw, Bill Gates and Boeing.
Certainly, Teledesic is ahead of the pack on the political/regulatory front. It is one of the first of the new breed of satellite providers to obtain both FCC and ITU frequency approvals and is planning to have almost a year's jump on global competition from Motorola's Celestri, a LEO/GEO hybrid system.
Teledesic plans to use its own fast-packet, ATM-like protocols as well as onboard processing and laser-based intersatellite links (each satellite is linked to eight adjacent birds). It will deploy terrestrial gateways to interconnect with other wireline and wireless terrestrial services. Teledesic also promises to partner with terrestrial providers, rather than offer direct services to businesses and consumers.
Teledesic President Russell Daggatt, says a key differentiation for Teledesic is the fact that it is not building its own technology but "assessing the state of the art and building a system around th
at." Daggatt says the challenges for Teledesic today primarily are those of execution on its plans.
While Daggatt says he "won't rule out" the possibility of providing services directly to large users "our thinking is to work through local service provider partners" and to be fully integrated with these providers from a technology and marketing standpoint.
The organizations he expects to line up first for Teledesic services are those with today's most common broadband needs-those requiring LAN-to-LAN interconnection, site-to-site interconnection, or home/remote access to the LAN. Daggatt says the ubiquity of systems like Teledesic will give it an advantage over terrestrial integrators because they can't provide integrated services in areas where high bandwidth doesn't yet exist. He also expects Teledesic to be widely used for wireless interconnection and backhaul.
Like many other satellite providers, Daggatt says broadband satellite will "never be able to match the economics of fiber wit
h heavy continuous traffic on dense routes." In other words in areas of population density or usage intensity it will be difficult to compete with terrestrial services. It's where those factors begin to drop off that Teledesic sees its opportunities. Daggatt also believes the unified infrastructure of broadband satellite systems will give it a quality of service advantage. In fact, he thinks QOS, "will emerge on the Internet as a bigger issue than bandwidth."
"It's safe to say pricing will drop to where it's a matter of per-megabyte pricing depending on QOS and that pricing will certainly be way below today's equivalent pricing," says Daggatt.
One reason for dropping prices is that bandwidth supply is on the rise-and Teledesic plans to be providing a significant share of that overall bandwidth (see chart). Even as it cut the number of its satellites from 840 to 288 Daggatt says overall capacity changes were subtle because the larger satellite systems being used support more capacity.
Wh
o does Teledesic see as its competitors? For real-time applications, Daggatt says he doesn't think GEO systems will "have much of a role, really any role, to play." He thinks GEOs will primarily serve up broadcast. He sees LEO competition eventually coming from Celestri and SkyBridge but says both systems have a long way to go to catch up to Teledesic's lead. Fiber and xDSL will compete, but only in the limited areas where they are deployed, he says. What about high altitude manned and unmanned craft like SkyStation or Halo? "Satellite technology is pretty straight forward," says Daggatt. "You have LEOs up there now for government purposes. It's been done. We know how to do it. It's a relatively low risk undertaking, where some of the blimps and things still have a lot of development risk." Daggatt says he doesn't want to dismiss such systems, "but we aren't factoring them in from a competitive standpoint."
|
REPORTS
ANALYTICS
Follow 
