Google is reportedly developing wireless networks for sub-Saharan Africa and Southeast Asia that would combine a technology well established for such purposes (TV White Spaces) with one that's a bit more exotic—balloons that transmit wireless signals.
The Wall Street Journal broke the story, reporting "the Internet search giant has worked on making special balloons or blimps, known as high-altitude platforms, to transmit signals to an area of hundreds of square miles." As we noted a week ago, Google reportedly "wants to connect people outside of major cities in sub-Saharan Africa and Southeast Asia through a combination of frequencies used for television broadcasting, special balloons or blimps transmitting on non-TV broadcast frequencies, and potentially through satellite transmissions."
Google itself is already delivering wireless Internet access to South Africa and other areas with TV White Spaces technology, so it makes sense to expand that effort. Satellite Internet is another technology well-suited to serve rural areas without strong wired Internet connections (despite latency problems).
So what about balloons? Is that for real or just another rumor based on unnamed sources that never sees the light of day? Google declined to say anything, noting to Ars that "we don't comment on rumor or speculation."
Speculation can be fun, though. Let's assume the Journal report is correct and try to figure out whether the scant details we have make any sense. Wireless consultant and engineer Steven Crowley told Ars that using balloons to distribute wireless Internet is certainly a possibility.
In fact, balloons are already used for various types of wireless communication. Some two decades ago, Crowley consulted for the US government, which was using balloon-based technology to transmit anti-Castro propaganda to Cuba through a US government-financed TV station called TV Marti.
After considering options such as flying an aircraft back and forth over the water or transmitting from a barge in international waters, the station went with an aerostat, a balloon tethered to the ground by a 10,000-foot cable, Crowley said.
"They'd feed up video and power on this cable and transmit to Cuba when the weather was good, which was most of the time," Crowley said. "At times they had to bring it down when weather was bad. I think that's true with all these systems. If you have inclement weather, depending on the design, you're going to have the wind blowing on the line or blowing on the balloon."
Balloon-based communications are often used today by the US military. A company called Space Data makes balloon-based repeater platforms for the US Air Force that "extend the range of standard-issue military two-way radios from 10 miles to over 400 miles." The balloons weigh 12 pounds or less and are designed for rapid deployment. Unlike the aerostat Crowley mentioned, this one isn't attached to the ground via a cable.
"Space Data’s military FM repeater technology is based on the simple concept of lifting wireless transceivers into the stratosphere (between 65,000 and 100,000 ft) using weather balloons technology," the company says. With transmit power of up to three watts and using frequencies of 225-375MHz, the technology can provide voice and data service to an area with a diameter of 400 miles. The balloon is controlled from a portable ground station. It can be tracked via GPS and has vent and ballast systems to control its altitude.
Balloon-based communications similarly are provided to the US Army for "encrypted communications for troops in the field." Beyond the military, a company called Oceus Networks was working with the Federal Communications Commission last year to demonstrate feasibility of high altitude balloons at "near-space altitude" to carry LTE signals to public safety responders in disaster areas. Oceus Network is also working with the Army to test LTE communications from tethered balloons much closer to Earth, just 650 meters from the ground.
Similar proposed models involve balloons tethered to the ground by a fiber optic cable connected to a base station.
From high altitudes, serving users from tens of miles away
For Google, Crowley thinks the most likely scenario is a cellular deployment using standard cellular frequencies and 3GPP2 specifications (rather than 4G) to take advantage of economies of scale from the wide deployment of older, cheaper handsets. "It depends on what the model is for handsets and devices, and getting these out to people," he said. "As LTE grows the device cost goes down and it becomes more practical to deploy."
With balloons rather than cellular towers, Crowley believes it would be more like a TV model where the transmitter can serve uses 30 to 50 miles way. "From a high balloon, you're going to Reach a little further beyond the Earth's curvature," he said.
The problem here would be more on the upload side rather than the download side. "You can receive a signal OK from that," he said. "The limit would be the uplink going back, how much power do you have from a small device going back and can you reach that? It's way up in the air. That helps. You get gain when you raise the height of the base station antenna, that makes it easier for the handset to reach."
Such a scenario would likely focus on delivering the most essential services, such as text messages, e-mail, and voice. "Even voice can be compressed to a very low level," Crowley said. "I don't think they're envisioning streaming Netflix or movies."
If balloons were used to distribute wireless signals, it would probably work best as a complement to other technologies, which is what it sounds like Google is envisioning with satellite and White Spaces networks. Google could provide both cellular connectivity and Wi-Fi-like systems. Wireless expert Peter Rysavy of Rysavy Research noted that he "can't comment on the feasibility of balloons," but told Ars that from an architecture standpoint "it makes sense for some scenarios to have base stations that have larger coverage areas than terrestrial systems, but smaller than satellite systems."
"The benefit I see is providing data service over a large coverage area, greater than you could with a White Space network," Rysavy continued. "The downside is that with such a large coverage area, you can’t offer that much bandwidth to each subscriber. For that reason, I don’t see it as that useful in the US except for rural areas."
Back in 1996, Wired wrote about how Sky Station International was planning to connect 80 percent of the world's population by 2004 from balloons 15 miles above the ground. Obviously, it didn't happen.
Crowley notes that while commercial wireless networks involving balloons have been considered in the US, when bean counters analyze the numbers the proposals tend to die. Even though you might need just one balloon where previously 10 cell towers were required, Crowley said companies doing the financial analyses have "reached the conclusion that for the US market at least, it's not practical."
Google may have motivations beyond finances, though. The company certainly does lots of charitable work. It has been working on improving connectivity in the US with Google Fiber and bringing the Internet to underserved populations overseas through White Spaces networks. So balloon-based networks might be an extension of Google's longstanding goal of bringing the Internet to as many people as possible, and it might help boost Android usage to boot.
As Crowley said, "maybe Google has another model for supporting this."