The return of the satellite constellationsby Jeff Foust
|
| “We’ve got to solve this [latency] problem if we want people to have dynamic HTML and web 2.0 applications,” Wyler argued. |
Today, constellations are back in vogue, in a big way. As the survivors of the 1990s boom and bust repopulate their constellations with next-generation spacecraft, new systems are planning to compete with them, as well as the geosynchronous satellites that remain the backbone of commercial satellite communications today. Will history repeat itself?
One person who hopes it doesn’t is Greg Wyler. He is a veteran of O3b, a venture that operates a small constellation of satellites in medium Earth orbit (at a rarely-used altitude of 8,000 kilometers) to provide broadband communications to underserved regions of the globe—the “other three billion” people from which the company takes is name.
Wyler is now working on a new satellite venture technically far more ambitious than O3b. Called OneWeb, it plans to place 648 satellites into LEO, providing broadband services globally. It sounds a lot like O3b or, for that matter, GEO systems that also provide high speed Internet access.
The difference, Wyler said in a speech at the Satellite 2015 conference in Washington March 17, is reduced latency, or the lag in transmitting and receiving data. LEO systems, being much closer to the Earth than MEO or GEO ones, should offer lower latency and thus improved performance.
“We’ve got to solve this problem if we want people to have dynamic HTML and web 2.0 applications,” he argued. Terrestrial systems can, of course, offer low latency, but can be difficult to install, he argued, recalling his experience prior to O3b trying to install fiber in Africa.
After months of speculation about Wyler’s plans—he was, at times, linked to both Google and SpaceX—OneWeb announced in January its plans to develop the constellation with the financial backing of Qualcomm and the Virgin Group. The latter would be offering launch service for at least some OneWeb satellites via Virgin Galactic’s smallsat launch system, LauncherOne, currently under development.
Wyler talked little about the satellite system in his Satellite 2015 presentation, focusing instead on the ground segment. He showed off several different kinds of terminals, including one that could be mounted on a roof and powered by solar panels; others could be mounted on aircraft or emergency vehicles, in the latter case providing seamless communications when out of range of terrestrial networks.
However, OneWeb is expected to move forward soon on the space segment. The company is reportedly weighing bids from five different satellite manufacturers in the US and Europe to build its fleet of satellites, with a decision pending as soon as next month.
| “I don’t think it’s that everyone else is doing something wrong,” SpaceX’s Shotwell said of competing satellite ventures. “We want to participate in this activity.” |
OneWeb’s announcement in January was almost immediately overshadowed by another, even more ambitious venture. Speaking at an event in Seattle closed to the media, SpaceX founder Elon Musk said he was in the early stages of planning a broadband satellite constellation that would eventually feature 4,000 satellites.
In a video taken by one attendee, Musk provided a basic overview of his plans, including the development of a satellite manufacturing facility in the Seattle area. He said he had filed paperwork with the International Telecommunication Union (ITU), which regulates satellite communications, for spectrum for his system, but didn’t offer many technical details.
Musk’s long-term vision for the satellite venture is to grab a small fraction of global Internet traffic—on the order of 10 percent—which, if successful, could generate billions of dollars of revenue for the company. Musk said that money would go towards his ambitions to develop human settlements on Mars.
At Satellite 2015, SpaceX officials didn’t offer much in the way of additional information about those plans. “I don’t think it’s that everyone else is doing something wrong,” SpaceX president Gwynne Shotwell said of competing satellite ventures during a March 16 panel session. “We want to participate in this activity.”
“It’s a huge market,” she said, citing the difficulties she’s encountered trying to get broadband Internet access to a ranch in central Texas. “It’s pretty clear there’s a huge need to expand capacity outside of huge population areas.”
In February, SpaceX announced that it had raised $1 billion from investors led by Google and Fidelity. At the time, most assumed that Google was investing to support SpaceX’s satellite constellation plans, given that Google had parted ways with Wyler’s OneWeb.
Shotwell, though, said that was not the case. “Google and Fidelity invested in SpaceX kind of for their own business purposes, not for the global Internet project we’re exploring right now,” she said in response to a question about it at a luncheon during the Satellite 2015 conference March 17.
OneWeb and SpaceX’s unnamed satellite system are not the only constellations out there. Another venture is LeoSat, which is planning a network of more than 100 satellites for broadband communications. Earlier this month, it hired as its CEO Vern Fotheringham, the former CEO of satellite antenna company Kymeta, the first public statement by the venture.
“We will be announcing some very significant strategic global alignments and service delivery partnerships over the coming months,” Fotheringham said in the statement.
This new wave of LEO satellite systems is not generating alarm among major GEO satellite operators—yet. In a panel session at Satellite 2017 March 17, the CEOs of the “Big Four” satellite companies—Eutelsat, Intelsat, SES, and Telesat—said it appeared history might be repeating itself.
“It feels like it’s repeating itself a little bit,” said Daniel Goldberg, CEO of Telesat. “There are two questions here. One, can you launch a system like that? Absolutely. Two, does it achieve an attractive return for investors? I don’t know. It’s not easy.”
Such ventures, he said, need technology improvements and reduced costs to build and launch satellites, two things he was skeptical could be achieved soon. “Today, I don’t believe there’s a viable plan where these satellites can be cranked out in sufficient numbers at low enough cost,” he said.
David McGlade, CEO of Intelsat, said he welcomed new entrants into the business, but wondered if the key advantage such systems provided—low latency—was really that important. “Does it matter that much for voice? Not very much any more. Gaming? That’s nice to have. 5G? Maybe,” he said.
He and other panelists argued that, when it comes to providing high-speed bandwidth, independent of latency, GEO satellites could compete with any LEO system. Several invoked one of the recent buzzwords of the satellite communications industry: HTS, or high throughput satellites, which are designed to provide broadband services from GEO.
“Nobody should underestimate the room for improvement GEO satellites have for the coming years,” said Michel de Rosen, CEO of Eutelsat. “HTS satellites have been a revolution we’re all taking advantage of.”
| “Today, I don’t believe there’s a viable plan where these satellites can be cranked out in sufficient numbers at low enough cost,” Goldberg said. |
Some of the CEOs, though, were aware that GEO satellites, and their companies, risked looking conservative and stodgy compared to the new satellite industry entrants. De Rosen went to comical lengths in an attempt to dispel that. “Image is something we can change rapidly,” he said, as he stood up near the end of the panel session and took off his suit jacket, tie, and dress shirt, revealing a plain black tshirt.
GEO satellite operators might need more than a change in wardrobe to adapt to a changing market, but they raise a good point: despite the influx of interest and investment, there’s no guarantee than any of these systems can close their business cases any better than the failed efforts of the 1990s. What’s different this time?
Barry Matsumori, a SpaceX senior vice president, recalled in another Satellite 2015 panel working earlier in his career at Qualcomm as Globalstar—a joint venture of Loral and Qualcomm—was getting started. “We didn’t take into account how fast terrestrial GSM was going to expand,” he said.
Technology, he argued, has advanced considerably since then, enabling more powerful systems. “You look at the technology that’s available now, the processing power, compared to what was available back then, and it’s night and day. There’s a big difference,” he said.
Others have argued there is growing demand for the broadband communications services that were still somewhat novel in the 1990s, along with advances in satellites and launch services. Together, companies believe that increase in demand, and decrease in costs, can make constellations both technically and economically viable now.
Who’s right: the skeptical GEO operators or the optimistic constellation entrepreneurs? The answer may come in the next several years.
