Facebook and announce flying internet


March 28, 2014

Facebook and are working on solutions to increase global internet access

Facebook and are working on solutions to increase global internet access

The Connectivity Lab team at Facebook has been working on new technologies to improve and increase internet access across the globe. Proposed solutions include solar-powered drones and geosynchronous satellites, both of which would be able to beam internet access to earth. The announcement was made yesterday by Mark Zuckerberg, on behalf of the partnership. was established last year with the aim of bringing internet access to the two-thirds of the world's population that doesn't currently have access. The partnership includes technology leaders, non-profit organizations, local communities and experts. Its founding partners include Ericsson, Samsung and Nokia.

Zuckerberg has been very much the spokesman for the partnership since its inception and, earlier this month, Gizmag reported that Facebook was in talks to buy Titan Aerospace, a company that manufactures solar-powered unmanned aerial vehicles. Yesterday's announcement reinforces Facebook's commitment to the cause.

The Connectivity Lab was begun using the same engineering talent behind Facebook’s infrastructure team and the Open Compute Project. In addition, the team comprises UK-based high-altitude long-endurance aircraft firm Ascenta, the Jet Propulsion Laboratory at NASA, the Ames Research Center at NASA and the National Optical Astronomy Observatory.

The team has been working on solar-powered high-altitude, long-endurance aircraft that can remain in the air for months at a time with a view to deploying them to provide reliable internet access to suburban areas in limited geographical regions. Low-Earth orbit and geosynchronous satellites are being investigated as a means of providing internet access to lower density areas.

For each of these approaches, the team is exploring the use of free-space optical communication (FSO), which employs invisible infra-red laser beams, as a means of transmitting data. According to, "FSO is a promising technology that potentially allows us to dramatically boost the speed of internet connections provided by satellites and drones."

In the video below, Facebook's Yael Maguire talks about the technologies that is working on.


About the Author
Stu Robarts Stu is a tech writer based in Liverpool, UK. He has previously worked on global digital estate management at Amaze and headed up digital strategy for FACT (Foundation for Art and Creative Technology). He likes cups of tea, bacon sandwiches and RSS feeds. All articles by Stu Robarts

It's good that they have money to burn because I think they are underestimating the difficulty of doing this. I don't really know what all their intentions are but I highly doubt its as simple as their mission statement.


I think a dirigible would be a better platform but it's not my money.

@ Diachi Where do yo see there being a problem? Solar powered plane? Silly but it has been done. Flying the drone? Yea right. Wireless internet? Even sillier.


I say the drone could be a good target to shoot down with the dragon plane they'll sell in the Hammacher-Schlemmer catalog!


@Slowburn Why is solar powered plane silly? Last I read on gizmag they had one that should theoretically be able to stay in flight for 5 years 24/7 with on-board batteries. If the maintenance costs and reliability can be made to match that of a satellite why not?

"Wireless internet? Even sillier. " Are you sarcastic here? Great idea no? Assuming radiation and noise can be dealt with or reduced..


Solar powered airplane is silly because of the cost and weight/power penalty but for the drone intended to fly for weeks at a time above the clouds it is the best generally available to non-government programs.

I should have written, "A silly thing. but it has been done." The "silly" on the second and third question was to the idea that that they are really problems.


@Slowburn There is a saying for identifying vaporware that is somewhere along the lines of "Does it depend on technology that does not yet exist?"

As of may of last year the world record distance traveled for manned solar powered plane was 950 miles. The record flight time for a UAV is about 2 weeks but keep in mind that they are not simply keeping the UAV in the air for the sake of being in the air. It will have to haul some potentially heavy and power hungry communications equipment and need to be in the air for months at a time. If they divert a portion of solar power to power the transceivers/equipment required or bring (heavy) batteries or both there are no easy answers. They may be above clouds but they still have day/night cycles and winter to contend with. Needing your day to day network to perform at lab conditions is dreaming, needing your day to day network to perform at multiples of world record attempts is vaporware.

Instead of paying for base stations and connecting them they want to planes to communicate with each other using lasers. The problem is light within fiber optics reflects off the outer wall of the fiber core staying within the fiber. There is a demo at the corning museum of glass that demonstrates this using water. A point to point laser used for communication must be line of sight from plane to plane and very accurate through things like air turbulence. Because the light would diffuse without a core to support it there would be much much higher signal losses without a fiber core so it forces the UAV's to be close in proximity, it forces the lasers to use an extremely high amount of power to overcome distance limitations absent the ability to have amps along the path like a normal fiber network would use to handle distance. As things get sloppy a lot of channel spacing and forward error correction is needed which will greatly limit the bandwidth available for the plane to plane trunk. Because of the harsher conditions of an open air laser each plane must do a full optical to electrical to optical conversion which requires more power than a simple optical pass through. It must receive the data, run a FEC, and regenerate it at each hop which will consume power. If there is a problem with one plane the other planes will need to close the gap or whole communication link is broken like Christmas lights with a bulb out. The actual communication stream with the ground will require RF signals that may need to be licensed from the FCC in the US and various other organizations around the world at great cost. The problem with using a special licensed spectrum/technology is the people on the ground may require special hardware to communicate with the network. With 3G/4G networks each antenna is connected to the Internet. With the plane to plane network the planes will have to carry all of the collective data for multiple other planes before it communicates the data back to a wireless ground station with connection to the Internet. Because that ground station may be through clouds etc. it has an entirely different set of problems communicating all that data to it. Lasers would likely be out and it would have to fall back to tried and trusted RF with limited availability and licensing and all. Remember it must have enough bandwidth to the ground station to support the combined network of several planes. What this means is many of the same limiting factors that limit other technologies still apply but now with an entirely new set of problems to boot. Again with the vaporware test: "Does it depend on technology that does not yet exist?" yes

So although solar planes have been done they haven't been done to that extent. An open air laser communications network is still a fairly new and difficult tech. That’s 2 major technologies the project depends on that don't really exist yet at least in that capacity. The other problem is like I mentioned ground stations are still needed and still communicated with via RF so the entire infrastructure of planes works as sort of like a network of range extenders for a RF ground station that still has all the same associated costs and limitations. Those same limitations exist as well between multiple users on the ground communicating to the planes themselves so you have 2 RF choke points and a lot of latency as well.

Their goal is to provide low bandwidth connectivity to places around the world but the Internet itself has changed. No longer is it about the transfer of text based information. Yahoo, Apple, and Amazon are making major video pushes and video itself is gravitating to greater and greater quality. All of this effort would result in a network that could barely sustain the Internet of today let alone 5 or 10 years from now. As other technologies move to things like channel bonding, DWDM, and multiple fiber cores to scale to increasing demand their network would largely stagnate due to several physical limitations. Even aside from some huge bandwidth bottlenecks the amount of latency involved in all the distance traveled before even reaching a ground station would make things like VoIP/VPN/online gaming difficult.

For the amount of effort needed to pull this off it would probably be both cheaper and easier to just license RF spectrum to expand coverage of mobile wireless networks in those countries. There are frequencies better suited to cover large geographic areas than the ones mobile carriers are using in the US and they are already using RF for 2 of the hops anyway.


@ Diachi

POWER SUPPLY: While for most purposes I don't like solar power and here I would prefer to generate heat by using a gamma emitter in tungsten container or a fission heat source and Stirling cycle engine but they are unavailable for real and imaginary reasons. So you are left with a ground based broadcast power or solar to make it work so we are stuck using solar as the optimal option. (That was a phrase I never expected to use.) However this does not necessarily mean photoelectric. Waste heat recycling may produce a significant portion of the energy as well given the temperature at altitude.

DESIGN OF DRONE: The biggest problem is energy budget and here the problem is in getting the drone to operating altitude with enough of a charge to get through the first night but sense batteries don't weigh more charged than drained the drone can use all the power from the solar collectors for the climb and if the drone takes more time than daylight is available for (A point I am not ready to concede) it can be launched before dawn using another plane to illuminate it with spotlights until the sun come up. Next comes carrying the electronics package unlike a man carrying plane the load can be distributed out through the wing reducing the structural requirements.

COMMUNICATIONS: Laser, The air at this altitude is both real thin and devoid of dust and moisture leaving it at least equally as transparent as optical fiber and sense the drone will have line of sight on each other their is no need to bend the light beam so no need for the reflectivity of the boundary layer in the optical fiber. Sense the drones will be constant formations the laser transmitters and receivers will not need a great deal of transverse to keep them on the target mostly just vibration dampening and a good telescopic lens on the receiver will greatly reduce the needed transmitted energy.

Yes they will have to buy the RF frequencies and build a ground station but the ground station can be located far out of town making the land prices low enough that acquiring it might be cheaper than acquiring the land for a single inner city cell tower and they won't have the problem of random vandalism.

There are technical challenges but absolutely no reason to think that they can not and/or will not be solved.


When I read an article like this, I begin to think. While technology is nice and has become a way of life for most of us, I find it very disturbing on many other levels. We are losing our freedom and privacy, were are becoming more dependent on big business and government to make major decisions and more vulnerable to whatever intentions they have. Our economy looks prosperous but that hides the fact that it is much more fragile than ever before. We are making ourselves much more controllable. We have become like flies caught in the worldwide web, unaware or even ignoring the hungry spiders. One big solar flare or well placed terrorist attack could bring it all tumbling down. It's not a matter of if these bad things will happen, it's a matter of when. We are rapidly reaching the point that we have become too civilized and technology dependent to survive. So what should we do? Keep ensnaring and seducing even more of the world's population into an unsustainable system? Our vices and desire for immediate gratification will always keep us from making the right decisions. Would I like to go back to the old days? No, but it would be the right decision. Maybe you agree with me or maybe you don't but at least try to be honest.


@Bob what exactly are you saying about internet access being freely available? How is that something we don't want? How does that limit our freedom?

Sammy Akharaz

@Sammy. The worldwide web will ultimately make everyone on it traceable, dependent and ultimately controllable.


I sort of agree with Bob that this is one of the downsides to this technology. Not only will Facebook know a ton about you but this would potentially enable them to know exactly where you are as well.

Nobody pays money to use Facebook, they are in the business of using data about you to serve hyper targeted ads and knowing where you are opens new doors. As a business how much would you pay per impression knowing a minute fraction of the people who see your ad will bite?

What if you could advertise to only people nearby who are interested in the types of products you sell? Who wouldn't be willing to do that? Take it another step, what if you could pay Facebook to learn about the people who surround your business?

Collecting data on you is what Facebook gets paid to do, allowing you to connect with and share info with friends is only the honey pot.

Post a Comment

Login with your Gizmag account:

Related Articles
Looking for something? Search our articles