The newest, fastest in-flight internet will launch in China

Want to connect in-flight to the latest in airborne wifi services? Starting in 2015 that will mean a trip to China. Air China has signed on as the launch/test partner for Honeywell’s GX Aviation connectivity platform. The system will be installed on Air China A330s starting in Q2 ’15. The service is based on the Ka-band satellite frequencies, similar to the ViaSat-operated satellites used by LiveTV today for their installations on JetBlue, United Airlines and Aer Lingus (only JetBlue has it available to customers today; the others are expected later this year).

Honeywell will be using the Inmarsat GX network which will include three satellites in operation by the end of 2014; the first was successfully launched at the end of 2013 and the other two go up later this year. These Inmarsat-5 satellites support spot-beam technology which increases the available bandwidth to a specific region while reducing the hardware required on the aircraft to handle the traffic.

Honeywell expects to deliver 50mbit download speeds and 5mbit upload speeds to aircraft around the world with the new constellation of satellites. This is an increase over even the high speeds in service today from LiveTV, though they say increases are possible without retrofitting the hardware on the planes. Similarly, gogo is promising comparable download speeds with their GTO offering which is expected to enter service in late 2014, though the gogo GTO service will remain viable only over the USA and Canada as it requires the terrestrial connection as well.

Coverage Map of the Honeywell/Inmarsat GX satellite constellation.
Coverage Map of the Honeywell/Inmarsat GX satellite constellation.

One big question facing Honeywell and the GX Aviation team is that of contention. The total bandwidth available in any one spot beam is limited and multiple planes in the same airspace will need to share. That’s not such a big problem early in the deployment cycle as there aren’t going to be many planes with the kit installed. But LiveTV and ViaSat saw the potential for such congestion as they planned for their implementation and ultimately waited until they had a second, more powerful bird in the sky before initiating service. The LiveTV solution is only viable over the continental USA right now due to the satellite coverage while Honeywell is aiming for global coverage to start. They have a fourth satellite in development as a hedge against failure of any of the initial three but also suggests that they can use it to augment coverage if the others launch successfully. But will that be enough? Global coverage is a key to providing the necessary service levels for international travel but it also means that some of the available bandwidth will be covering portions of the globe with minimal demand while higher demand regions may not have enough bandwidth available. Looking at the map above suggests that coverage over the USA comes from about 5 spot beams; the same number cover the Brazilian Amazon region.

Even though the Ku-band satellites in service today have lower total bandwidth available and higher per-byte costs they carry the advantage of having more in orbit. This arguably means less contention on any single bird, though that’s mostly a theory since not all of the satellites are available to all the carriers. ViaSat will be launching their ViaSat-2 Ka-band satellite in mid-2016 which will significantly increase capacity and reduce per-byte costs, but it will still have a much more limited coverage area.

ViaSat-2's coverage map; the satellite will be launched in 2016.
ViaSat-2’s coverage map; the satellite will be launched in 2016.

So, who wins at the end of all of this? Arguably the passengers as they continue to see increases in coverage areas and bandwidth available, plus costs should continue to drop as the more efficient satellites take to the skies. That’s a very good thing. And given that some airlines are still deploying systems 2+ generations old, with high costs and low performance it should be a most welcome development to have the better service offerings in the sky. But there’s also the challenge for the airlines of getting something installed and functional to meet the demands of their customers while considering the future needs. JetBlue and United both took heat from their customers for not having wifi on board sooner (though it is not entirely clear what impact this had on booking revenue). And installing a system today typically means a multi-year commitment to justify the operational certification process and capital expenses of acquiring the kit and installing it on the aircraft.

As an airline do you wait for the Ka global coverage late in 2015, knowing that you won’t have the hardware on board before 2016 at the earliest? Or do you make a move now with a lesser technology platform? Doing both is likely far too expensive a proposition, particularly in an industry where it isn’t clear what the yields are from having such a service on board in the first place.

Never miss another post: Sign up for email alerts and get only the content you want direct to your inbox.

Seth Miller

I'm Seth, also known as the Wandering Aramean. I was bit by the travel bug 30 years ago and there's no sign of a cure. I fly ~200,000 miles annually; these are my stories. You can connect with me on Twitter, Facebook, and LinkedIn.


  1. Very interesting. Currently the Chinese airlines go nuts about taking out phones in airplane mode on flights, though allow computers and tablets. Now they plan to leapfrog everyone in connectivity.

  2. “Inmarsat-5 satellites support spot-beam technology which increases the available bandwidth to a specific region while reducing the hardware required on the aircraft to handle the traffic.”

    – how is the hardware required reduced? Less components/LRUs? Less installed system weight?

    LiveTV/ViaSat is 12Mbps to individual pax (which have been proven to be true at least on the 1st media trial, as everyone one the plane was able to stream. Be it, it was the 1st airplane with only 40-50 people using it). GX is ” up to” 50Mbps to aircraft. I am sure neither the ” to the aircraft” nor ” to the pax” is “guaranteed”! So in the end it comes down to what the supplier legally contractually committed to.

    1. The spot beams allow for a smaller antenna on the aircraft receiving the same throughput.

      LiveTV/ViaSat is NOT 12Mbit/passenger. That was debunked (by me) a while ago. It is still based on an aggregate total bandwidth to the plane and peaks at ~30Mbit today given the current settings on the kit installed on those planes. It can be adjusted up to ~80Mbit with software-only changes (no retrofit/STC work required).

      The 50MBit number has me skeptical given contention and such both within a spot and for the overall satellite based on the Inmarsat global constellation running with only 3 satellites. But we’ll see.

      I am quite confident, however, that Ka-band service has higher peak speeds available to the aircraft at a lower price point than Ku or L band options. In that sense this will absolutely be a faster option for airlines and consumers.

      1. I am /was involved in the connectivity evaluation at an airline (as an employee /contractor) and can tell u that LiveTV system weight is not lower than Ku on SWA or UA.

        As far as Ku performance comparison, you are right when comparing to wide beam Ku vs ViaSat. Ku (for e.g.Epic will have smaller spot beam than GX and can aggregate beams) spot beam will have higher throughput than GX. ViaSat Ku will likely remain slightly better as it has smallest of spot beams.

        Yes- it is all about size of beam and how the antenna is designed (although most have access to same antenna physics more or less…it comes down to manufacturing)

        1. GX capacity will not be dedicated to commercial airlines only, it will be shared with non aviation users

          1. Same is true with the ViaSat Excede product and pretty much everything else out there today. It is all shared. But the much higher overall capacity gives a much stronger potential for higher bandwidth to each plane than the prior generation solutions.

Comments are closed.