Commentary | Satellite Glamour Feeds the Industry


Expansive ideas for using satellites have excited entrepreneurs and investors for more than 50 years. Tens of billions of dollars have flowed into vast space projects that have had long-lasting impacts on global technology. 

The Apollo project to put men on the Moon stimulated technologies to reduce the size of electronics. Many people attribute the development of transistorized radios to that era. Computer technology and software were driven by the requirements to achieve the first landing on the Moon.

The U.S. Congress created Comsat Corp. as a monopoly in 1962. Its initial public stock offering was wildly popular and oversubscribed. From 1965 to 1980, Intelsat and Inmarsat laid the foundations of fixed and mobile satellite communications under the protection of international treaties. Eventually they became private companies. Enthusiastic entrepreneurs started Galaxy, SES Astra and PanAmSat. 

The first huge commercial satellite success came from the innovation of direct-to-home satellite television broadcasting led by DirecTV and Dish starting in 1993. These systems have invested about $5 billion combined for the space segment. They now serve more than 34 million subscribers in the United States alone, and those satellite operators earn annual revenues exceeding $45 billion. Globally, satellite direct broadcasting is a major industry.

When cellular technology became a reality in the early 1980s it was obvious that cellular towers would not reach everyone. Satellites could augment delivery to unserved regions. Companies in the United States formed the American Mobile Satellite Corp., and Telesat Mobile Inc. was established in Canada. These companies wrote business plans to serve 600,000 subscribers in remote regions and invested about $2 billion. When they launched in 1994 the market had changed and they served only a tiny fraction of the expected market. Nonetheless, that business evolved through new ownership structures and eventually became Terrestar and LightSquared. The new owners invested several more billions of dollars in satellites that were to be augmented with terrestrial repeaters. These systems were designed to provide service to hand-held telephones. Unfortunately, the latest incarnations have not been able to serve a significant number of subscribers either. Hundreds of employees were well compensated for building and operating these systems.

By the early 1990s Motorola had been highly successful in the cellular industry and concluded that low Earth orbit (LEO) satellites could provide low-latency communications to personal telephones. Iridium was developed with a constellation of 77 satellites and ground infrastructure that cost more than $5 billion. At the same time Globalstar was created by Loral with a constellation of 48 satellites that eventually cost over $5 billion. Both of these systems expected to serve several million subscribers globally. When they were launched between 1999 and 2000 the users did not materialize to support these projects. Iridium filed for bankruptcy in 1999; Globalstar did the same in early 2002, and its parent, Loral, filed for bankruptcy in 2003. A third system, ICO, initially sponsored by Inmarsat, faced a similar fate before it was fully deployed. Nonetheless, both Iridium and Globalstar emerged from bankruptcy, and they have been providing telephone service, paying employees, and even borrowed money to launch second-generation systems. After 15 years, all of these mobile satellites together are serving fewer than 1 million telephone subscribers. The U.S. Defense Department has made extensive use of their services in the Middle East wars.

By 1995 the Internet had become a new sensation. Teledesic was introduced as a fantastic constellation of 840 LEO satellites that would provide low-latency, global broadband service to unserved regions. Bill Gates of Microsoft and Craig McCaw, a pioneer in cellular communications, backed the project. The International Telecommunication Union allocated Ka-band spectrum specifically for nongeostationary orbit (NGSO) fixed services. Manufacturing companies that expected to build the system funded the early development work. As the space segment cost estimates soared near $20 billion, the constellation architecture was redesigned with fewer satellites. This process was repeated with smaller and smaller constellations until the project was dissolved in 2003. Another large obstacle to Teledesic was the high cost of terminals with two antennas that had to track moving satellites (both the departing and arriving satellites.) 

The excitement about Teledesic and the need for Internet access in remote areas led to an avalanche of dozens of other NGSO solutions. Alenia proposed an NGSO system called Skybridge that used Ku-band on a secondary (noninterference) basis. In case of interference between Skybridge and existing geostationary Ku-band systems, the Skybridge system would cease transmission. Skybridge, like Teledesic, expected its suppliers to invest in the project.

Lockheed Martin proposed and began to build an extensive constellation of several geostationary satellites with intersatellite links called Astrolink. 

All of these Internet access initiatives were terminated after billions of dollars were spent. For several years, hundreds of engineers had interesting and lucrative jobs exploring sophisticated constellations.

Finally, in 2005 and 2006, Spaceway and WildBlue, two geostationary Ka-band systems, emerged, and they are providing broadband services today. EchoStar/Hughes uses the Spaceway-3 and Jupiter-1 satellites, and ViaSat uses the WildBlue-1 and ViaSat-1 satellites. These two systems have more than 1.55 million subscribers in the United States and annual revenues of $1.7 billion. These systems appear to be profitable businesses that are providing broadband services cheaper, better and faster than some terrestrial alternatives.

A decade after Teledesic was terminated, a company called 03b Networks developed a constellation of NGSO Ka-band satellites that orbit around the equator at an altitude of 8,063 kilometers. These satellites have multiple spot beams that focus high-throughput communications to Internet service provider in remote areas. Each provider needs at least two large and expensive antennas that track the satellites as they pass overhead. This system is ideal for island nations in the Pacific Ocean that don’t have access to fiber networks. The first four O3b satellites are insured for $318 million, about $80 million each.

The most recent thrilling news has been the announcement that Google plans to pump billions of dollars into WorldVu, a constellation of LEO satellites similar to Iridium’s but operating in Ku-band. There would be 180 to 360 satellites so that the operating satellites would appear relatively high in the sky. This would enable the use of sophisticated electronically steered user antennas. The satellites would have satellite-to-satellite links like Iridium’s to form a network with a small number of gateways. The WorldVu satellites are likely to be more capable and expensive than the O3b satellites since they would provide broadband services to millions of individual subscribers.

Of course, the WorldVu system would face a wide range of technical and cost challenges. It would need to devise a way to provide high availability while protecting existing Ku-band geostationary satellites. Most of the satellites would be over ocean areas where there are few potential subscribers most of the time. 

New technology can be used to reduce costs and improve performance. Furthermore, Google would provide the financial fuel to keep an army of engineers, scientists and technicians employed for many years. 

Congratulations to the Google team for joining the ranks of those who are willing to provide the resources to infuse new vigor into the satellite industry.


Roger Rusch is president of TelAstra Inc.