by Theresa Foley
The coming wave of Ka-band satellite systems promises to alter the economics of the satellite business radically. Some of the new systems will have technologies that will enable digital traffic to be packaged in a way that provides operators with as much as 10 times more revenue per spacecraft than today’s satellites.
As Ka-band concepts mature, it is clear the term means different things to different people. While some Ka-band projects will be relatively rudimentary "bent-pipe" satellites that simply relay a signal, the systems now being designed by Astrolink and Spaceway will be a whole new breed of spacecraft.
Celso Azevedo, president and CEO of Astrolink, the GEO broadband system backed by Lockheed Martin, TRW and Telecom Italia, says, "At the end of the day, the Ka-band satellites can have 20-25 times the capacity of Ku-band bent-pipe satellites. Each Astrolink satellite will have the potential revenue of as much or more than the entire fleet of Intelsats." Intelsat’s current 20 or so satellites earn about $1 billion a year in revenue.
The increased capacity is the result of several factors: more spectrum, more frequency reuse and onboard processing that will allow operators to load several customers onto the same circuit. The Ka-band satellites will cost twice as much to build as the simpler Ku-band ones because of the new technologies and large size of the spacecraft. Customers will pay 5 to 50 cents per megabit for usage, compared to 20 cents to $1 for similar terrestrial service, Astrolink says. Data rates will be 400 kbps to 20 Mbps.
Several techniques will be used to make Ka-band spacecraft more profitable than traditional satellites. First, the higher frequency allows a narrower beam to be used to reach the same size dish as Ku. Narrower beams mean the frequencies can be reused more times, providing about six times the capacity of Ku-band bent pipe satellites. Each Astrolink satellite will have 58 beams, and frequencies typically can be reused in every other beam, as long as adjacent beams don’t use the same frequency. Designing the satellites with multiple spotbeams and onboard processing that multiplexes several customers onto a circuit provides three to four times the efficiency of simple Ku-band satellites as well.
Astrolink has 60 percent of the money in hand for its first four satellites. This money comes from Lockheed Martin, TRW and Telecom Italia, with their 46, 27 and 27 percent respective ownership stakes. Astrolink’s first satellite, for service to all of the Americas, should launch by year-end 2002, with service in the first quarter of 2003. The other three satellites will serve Europe, the Middle East and Africa; Asia; and the Atlantic Ocean region, in that order.
Although numerous companies are planning to launch Ka-band or broadband satellite projects, Astrolink’s closest competitor in the category of highly sophisticated geostationary Ka-band systems now appears to be Hughes’ Spaceway project, which has 100 percent of its financing in place for the first phase.
Spaceway has a $1.3 billion commitment from Hughes to pay for its first three satellites, which Ed Fitzpatrick, Spaceway’s vice president of business development, says gives Spaceway a huge advantage over the competition. Spaceway will launch its first satellite in 2002 with service to start that same year.
Spaceway is being promoted as cost competitive with some terrestrial broadband networks, based on the improved economics of a satellite with the frequency reuse abilities of the new system. Fiber cables will be a better buy for customers who want to "constantly pump" their traffic at the same rate, Fitzpatrick says. But, he says, for customers who want "bandwidth on demand"-a pricing option that Hughes’ officials say will become increasingly important-satellites can beat fiber.
Each Spaceway satellite will have more than 100 spotbeams, enabling 25-fold frequency reuse. Hughes will incorporate a technical advance called a fast packet switcher-router, which describes the onboard processing system used to take packets of data and route them from one beam to another at speeds of 10 Gbps, to get the maximum traffic onto those beams and increase their value.
Spaceway will decide in early 2000 whether or not to include intersatellite links to route traffic from one region to another on the global system, or simply tap into terrestrial networks such as undersea cables to route the intercontinental traffic. The trade off is a business one, based on how much it will cost to add the ISLs and how much intercontinental traffic the Ka-band systems will be projected to carry.
Meanwhile, the non-geostationary Ka-band satellite project planned by Teledesic appears to have fallen behind the geostationary ones by two to three years. Teledesic will be ready to disclose technical details of its $10 billion-plus non-geostationary Ka-band satellite system around mid-October, following the signing of a "systems agreement" with prime contractor Motorola Satellite Systems on July 9.
Teledesic co-CEO Bill Owens says the agreement is the same as a firm, fixed price contract. He says the satellites will have optical intersatellite links and a lot of new software. Teledesic officials are hoping that the system’s lower altitude orbit and the correspondingly shorter signal delay in transmission will give it an advantage for point-to-point interactive traffic, but the company also plans to offer point-to-multipoint services and should be competing with the geostationary satellites in many respects.
Theresa Foley is Via Satellite’s Senior Contributing Editor.
