By James Careless

When it comes to satellites and the Internet, the news is both bad and good. It’s bad for those offering interconnecting terrestrial Internet backbones, or other point-to-point applications. But it’s good for those carriers offering point-to-multipoint satellite Internet delivery.

To put it another way, “satellite is not good for point-to-point, unless it’s the only option available,” says Stephen Blum, president of Tellus Venture Associates. However, “For point-to-multipoint it’s a killer. You can’t beat satellite.”

How can two such contradictory views apply to the same telecom sector? Easy. Point-to-multipoint services exploit satellite technology’s unique edge: they cover entire hemispheres from a single point in space, and use that position to serve billions.

However, point-to-point satellite links merely mimic the capabilities of terrestrial fiber optic networks. Given the explosive growth of those networks–especially transoceanic–satellite point-to-point Internet links are finding themselves increasingly under siege.

Point-To-Multipoint: Looks Like Good News

When we speak of point-to-multipoint satellite Internet service, we’re generally talking about those carriers who link customers directly to the Web. Hence, we’re talking about satellite ISPs operated by companies such as Hughes Network Systems (HNS) and Gilat.

To say the least, both these companies have been busy staking claims to cyberspace.

For instance, in January 2001, Hughes announced its “DirecTV Broadband Powered by DirecPC” consumer product. Aimed at the company’s existing nine million DBS subscribers, “Powered by DirecPC” provides two-way high-speed Internet access by satellite, with a downstream speed of 400 kbps and a return path of 128 kbps. To sweeten the deal, Hughes is also packaging over 225 DirecTV channels, and streaming it all through a single “DirecDuo” dish at each receive site.

“This is a key element in the Hughes strategy to bring ‘infotainment’ convergence home,” says Paul Gaske, HNS’ executive vice president and general manager, when the new service was announced in Las Vegas. “With the addition of DirecTV to our other ‘Powered by’ partners [such as America Online, Earthlink, and Juno], DirecPC will now be marketed to a combined base of over 45 million existing subscribers of television and Internet services in the United States.”

Meanwhile, Gilat’s Starband Communications passed the 25,000 subscriber mark on March 12 of this year. Given the service was launched a few short months ago, this is quite an accomplishment, and one that Starband CEO and co-chair Zur Feldman exults about.

“In just over a year, Starband has gone from a business plan to over 25,000 subscribers, national distribution channels, and an infrastructure to support the demand from consumers still waiting for high-speed Internet access,” he says. “We will continue to execute against a key company objective: to bring high-speed Internet access to virtually everyone, everywhere, in the United States.”

Is all this effort justified? If satellite ISPs perform as well as DBS has to date, frankly yes.

According to Blum, U.S. DBS penetration will hit 18 million subs this year and 36.1 million by 2010. That’s one out of out of every three households receiving TV from space, on a platform that–as Hughes has correctly deduced–can carry high-speed Internet as well.

Making matters better–and how often can one say that these days?–is that DBS subscribers not only earn more money than the general population–a median average of $48,200 versus $38,885–but are also the kind of consumers about whom advertisers fantasize, according to Michael Goodman, a senior analyst with the Yankee Group.

New DBS subscribers are even more attractive, Goodman adds. They’ve got a median income of $56,800, and, on median average, are 41.4 years old.

Point-To-Point: The News Looks Bad

At first glance one would expect point-to-point satellite Internet to be as promising as point-to-multipoint. After all, “today the number one opportunity is still Internet backbone connectivity,” says Gina Dolan, Intelsat’s senior product manager for Internet.

So what’s the problem? Two words: terrestrial networks. Armed with ever-increasing pots of fiber optic-based bandwidth, the world’s terrestrial carriers are in a position to cut prices, and make satellite point-to-point’s future very dim indeed.

To put it bluntly, “fiber is reducing opportunities in the core business,” says Bert Liebowitz, NetSat Express’ former president and COO, and now a consultant.

Greg Caressi agrees. A research manager for Frost & Sullivan’s Telecommunications Services Group, Caressi recently told delegates to the SATELLITE 2001 conference that “where satellite has been used to trunk Internet traffic in metropolitan regions –particularly in China a few years ago–it’s already been replaced by fiber.”

To make matters worse, it’s not just inter-city fiber capacity that’s growing. It’s also international capacity, as submarine cables reach across the oceans, like a fiber optic web strangling satellite traffic at the neck.

How big is this growth? Well, within the next two years, there will be about 9,000 Gbps’ worth of terrestrial bandwidth available in the South Asian region, says Stephan Beckert, TeleGeography’s director of traffic research. Meanwhile, about 7,000 Gbps of fiber will connect the United States to Europe, and 6,500 Gbps will run between the United States and Asia.

To put things another way, just one of these new cables can carry more traffic than the world’s entire satellite fleet combined. Small wonder that those in the know are worried about satellite point-to-point’s viability; especially after what’s happened to international point-to-point telephone traffic by satellite. Once the workhorse of this sector, traffic has declined so much that there are no longer any such links between the United States and Germany, says Beckert, nor the United States and Japan. His conclusion? “The point-to-point era for satellite communications has come to an end.”

While all this is going on, terrestrial wireless carriers are also hemming in satellite companies, says Jean-Francois Gambart, Alcatel’s vice president of marketing and business development.

If terrestrial carriers are allowed to continue grabbing spectrum–notably for 3G wireless and other broadband applications–then Gambert fears the satellite industry’s future will be put “at risk.”

Point-To-Multipoint Has Its Internet Challenges Too

Before you ditch all your Internet point-to-point traffic in favor of point-to-multipoint, you should know this latter application has its problems too–namely capacity.

To illustrate this point, Blum points to Starband. Based on his data, Starband needs to accommodate at least 20,000 subscribers per transponder to make a profit. Right now Blum estimates the company’s got about 8,500 on each.

Of course, these are early days for Starband; given time, it will likely hit the mark. But what about broadband? Well, that’s where things start to fall apart, Blum says.

For instance, say that Starband users start to download 220 kbps video streams: the kind already being delivered by PBS. According to Blum, a single transponder can only support about 200 such users. Even if they turn off the video and switch to 16 kbps streamed mono audio, the most each transponder can support is 3,000 users.

The result? If it proves popular–and it already has–broadband Internet traffic could hit satellite ISPs with the double-whammy of lower user-capacity and higher per-user cost. For a nascent industry, this is just plain bad news.

What Can The Industry Do?

Given the historical decline of point-to-point satellite traffic, and the capacity problems looming for point-to-multipoint, satellite carriers are faced with some hard choices with respect to the Internet. If they commit themselves too much to this medium, they leave themselves open to the ever-increasing strength of terrestrial competition. But if satellite carriers sit on the fence, they risk losing this battle before it even begins, by surrendering the field at the outset.

So what’s the answer? If you can’t beat ’em, buy ’em.

By buying terrestrial networks, satellite carriers can play both sides of the game. They can deploy satellites in emerging markets where nothing else exists. Then, as competing terrestrial networks begin to extend their tendrils into the market, satellite carriers can move their traffic onto their own fiber optic networks. This allows them to compete head-to-head on cost and technology, while freeing up their satellites to tackle new opportunities.

But that’s not all a combined approach offers. By using both satellite/terrestrial networks for Internet service, carriers can direct traffic based on which route makes the most sense. Are 500,000 users trying to simultaneously access the latest Victoria’s Secret Webcast? Put it up on satellite. Is most of your traffic moving within certain localized areas? Save the transponders: put it on fiber optic lines.

This truth has already been grasped by Bell Canada Enterprises (BCE). Owner of Canada’s Bell ExpressVu DBS service and Bell Sympatico’s high speed DSL platform, BCE recently announced that it was integrating the two in a single set-top “ComboBox.” It’s being built by Echostar Communications, which is already Bell ExpressVu’s prime receiver supplier.

By integrating its satellite and landline networks, BCE will enjoy the very efficiencies outlined above, says Tom Hope, the CTO at BCE subsidiary Bell Canada. But that’s not all: thanks to the ComboBox’s planned 40 Gb hard drive–one already found in the Echostar PRO 501 set-top box–this carrier has also figured out “how to provide Video on Demand,” he says.

“Everybody has been thinking about network storage and then immediate delivery through a transport network,” Hope explains. “Well, the ComboBox allows you to download the top six movies [directly to the box via satellite], and they sit there [in the set-top’s hard drive] encrypted and secure. Then, if the user wants to watch them, they play immediately and with full VCR functionality.”

Will Customers Accept Integrated Solutions?

In a word, yes.

This is because people “don’t care how you move their traffic,” says Kay Sears, Verestar’s vice president of voice data and Internet services. “They just want a good connection to the Internet.”

Given this very simple fact, satellite industry experts are “predicting convergence on a service level, rather than on a technology level,” says Blum. “What you’re going to see over time are hybrid networks that use terrestrial for point-to-point, and an overlay of broadcast broadband coming in via satellite.”

Beckert agrees. In fact, he believes it’s starting to happen right now. After all, “we’ve all heard announcements of Intelsat or Panamsat becoming a communications solutions provider,” Beckert says, “rather than a satellite company who only sells transponders.”

Tackling The Future Successfully

Clearly, the approach being espoused here runs counter to satellite industry tradition. In fact, the last 50 years have seen people pitting satellites versus landlines in a kind of transmission “Holy War,” as if respect for one meant disdain for the other.

The satellite industry can no longer afford this kind of thinking if it’s serious about the Internet as a long-term proposition. The development of international voice telecommunications is a warning: where satellites pioneer, terrestrial networks profit. The only way to win is to be in both.

Of course, satellite carriers can ignore this advice. In doing so, however, they’ll find themselves more and more squeezed out of the best markets–namely first world urban centers–and increasingly dependent on the worst: namely the rural third world.

This isn’t a value judgement; it’s just that the first world has more money to spend on telecommunications. Unfortunately, the third world doesn’t: that’s why rural satellite telephony has yet to live up to its potential.

It’s also why Iridium and Globalstar found themselves in such deep trouble. By the time they launched, terrestrial competitors had skimmed the rich cream of the wireless market. Granted, you still can’t get cellular phone service at sea, at the North Pole, or in the Australian Outback. But just how money is there to be made by supplying satellite services to these regions? Not much. At least not enough to keep most for-profit businesses alive.

This is why, for point-to-point and point-to-multipoint satellite Internet providers alike, it’s time to take a hard look at adding terrestrial networks. These days, satellite alone just isn’t enough.

James Careless is a contributing writer to Via Satellite.

---