PARIS — Airlines and in-flight connectivity service providers say they want hardware to be interoperable. Why is it so hard?
Astronics, Boeing Global Services, Newtec and Zodiac Data Systems agreed: Standardization of in-flight connectivity hardware so far is going nowhere, except maybe through the rare M&A such as ST Engineering’s bundling of Newtec with VT iDirect.
Addressing Euroconsult’s World Satellite Business Week here Sept. 9-13, these companies offered a ground’s-eye view of why, in addition to the lack of standardization, none of this — installation, regulation, life cycle, future-proofing — is obvious, or inexpensive. As will be clear in the discussion, neither Viasat nor Hughes were present.
A 4-pound (1.8-kilogram) bird and a radome
Stephane Bloch, director of international operations, Astronics: For large radomes or anything external, there is the bird strike requirement where you have to show that if your system is hit by a bird of more than X pounds, the aircraft will survive. For engines it’s birds over 8 pounds, and for whatever reason for radomes it’s birds over 4 pounds. So you have to run all the tests, you have to shoot actual birds in the radome.
[U.S.]FAA and [Europe’s] EASA have this interesting game of whenever someone puts out a requirement the other one makes a best effort to put up another requirement slightly different from the first. So one of the biggest constraints is how do I design something that will match all the FAA and EASA requirements, and how do I design something that will last throughout the years, not in terms of aging but in terms of allowing with the highest possible probability the operator to upgrade his systems.
One of the industry answers to that is ARINC 791 [Ku- and Ka-band aeronautical terminal equipment standards], which was supposed to provide interoperability. It did not. For example, the involvement of Boeing in ARN 791 was limited, or it did not have the same interests as Airbus. Our friends from Panasonic did something slightly different than Viasat, etc.
‘Is the IFC ecosystem moving to interoperability? No’
Everyone speaks about ARINC 791 but I’m not sure a lot of people are actually following it. So do we have a real industry normalization? No. Does the industry have a clear path to preserve the future in terms of systems being interchangeable and interoperable? The answer to that is clearly: No.
We are not yet at the transition between gimbaled antennas and phased-array, but we can see that coming. How do we design something now that will eventually allow phased array antenna to be installed on the aircraft that will have previously been severely modified if not damaged by the installation of the current generation? Yet can we wait for the next generation? No, our industry would die.
Brian Saunders, strategy lead, Boeing Global Services: On the lack of standards: Putting out aircraft at rate, you don’t want the airplane varying as it comes down the line – this one’s going to have Panasonic, that one’s a Gogo system. Even in the retrofit market, the variations across all those and the FCC processes that go into that, it’s something that needs to be dealt with.
We can actually learn a lot looking at some of the standards brought into the cell phone industry. It’s different for integration onto the aircraft, but really driving toward standards and then making people stick to those standards. Because each of the providers wants to tweak their solution in such a way that they have an edge over the other guy.
Q: How do you feel about standardization, different technologies and slight variations? For your IFC business do you have one modem or multiple modems?
Thomas Van Den Driessche, chief executive, Newtec: We have a third-generation modem, one that actually works. We have a couple of thousand installed and we’re peaking at 50 installs a week. You may be familiar with the Panasonic network we did. We’re standardizing by consolidation [Newtec was recently purchased by ST Engineering, which also owns VT iDirect], by patching different satellite networks together. We do have a single network over an SES high-throughput fleet, and Eutelsat and Intelsat and regional Ku-band satellites – patched all together. We get close to seamlessly switching between them. Newtec by itself is actually a standardization company, we are in multiple standardization bodies.
‘Don’t blame us modem manufacturers’
The industry is not great at standardization because they look at the modem manufacturer to standardize. If you compare it with the cellular world, it’s an ecosystem problem, not a modem-manufacturer problem. Our modems are software defined so we have no problem running standardized stuff. An aero modem is a software-defined modem these days. You can upgrade it over the air even, although you’re not allowed by FCC rules, that’s another thing. So looking at modem manufacturers is the wrong thing.
Look at all the different payloads there are. Every satellite operator is now on the verge of coming to the new digital payload that does something completely different than the next MEO or LEO guy, and every payload is different. That’s one reason why it’s become very difficult to standardize. It first starts with the antenna choices, is it Ku, is it Ka, is it combined, can we really combine that on an airplane? The ground segment will not be the problem, it’s the other part of the ecosystem.
Q: What is the biggest pain point today in IFC integration
Jean-Marie Betermier, president, Zodiac Data Systems: The industry is still very much suffering from the silo organization. Service providers are in silos trying to differentiate themselves by adding a new feature. We know the technology is here now, but this verticalization doesn’t help the final decision-makers when it comes to investing a lot of money for a minimum of five or six years. Software defined modems are now reality. What prevents us from delivering a software modem with some more standardization? I’m not talking the same waveform, the same access, but at least having hardware that could be installed in aircraft.
Van Den Driessche: The modem is indeed not difficult, today we can provide modems that can do that. If you can get the rest of the ecosystem in line with that, that’s the issue. For instance, there’s a difference between a Panasonic network and a Viasat network – Viasat is completely vertically integrated. We can provide a modem that can be software-upgradeable but I’m not sure Viasat would actually want it. Today it’s possible to put the same modem on every airplane, but it’s not possible to put the same antenna on every plane,
Q: Integrating equipment, has it gone down in terms of pressure, or is it just cost?
Bloch: Having agnostic modems would be a great step forward. We’re talking about satcom, satellite connectivity, IFC, but IFC is usually a complement to something inside the cabin. No one will install a very high-level IFC with no IFE system inside the cabin. Then you need to consider a server, CWAPs [cabin wireless access points], etc. Efforts are in process to provide the next generation of equipment that is fully agnostic and interchangeable, instead of having one server and several CWAPs, for example.
‘Changing broadband is peanuts, technically. But antennas…’
Swapping your broadband is technically peanuts. Financially, it’s a lot, it’s an investment. The big issue will always be the antennas. Swapping an antenna is a big thing. Changing the generation of an antenna may be a major undertaking and in some cases may be close to impossible. In some extreme cases your only option is to risk the aircraft and start from scratch, which is basically a $1.2 million ticket per aircraft. The message is the modem or any airline interchangeability will never be an issue from a technical standpoint, the big thing will be the antenna.
Q: Do you think if the cost for equipment will pick up? What is the impact in terms of adoption by airlines?
Saunders: Airlines want to make that investment once and have it pay off for a long time. They are saying ‘I understand I need this on my aircraft, but I don’t want it to have all these variabilities in it, I want to make the investment once and I want it to last for a relatively long time.’
Ku vs Ka: Pick one
Van Den Driessche: The modem can handle GEO, LEO or MEO types of constellations. The actual constellation shape and number of satellites doesn’t really make a difference. In the end, the difficulty is still Ku-Ka. The rest you can solve. Handovers, LEO/MEO, timing, schedule, synchronization, Doppler speed, software-defined radio, that’s all solvable. The difference between Ku and Ka is the big issue for airplanes. And there are no good solutions that have both on them to install.
Betermier: You hit the nail on the head. Just take a look at the new NGSO constellations [like] OneWeb and Telesat. They will both offer mobile connectivity, and more specifically to aircraft. One will operate in Ku, the other in Ka. What type of bet are you ready to make, even if you believe the NGSO will bring a lot of advantages? It’s very scary.
When we launched the first development three years ago, we decided to go for the Ka because the higher the frequency, the higher the bandwidth, the lower the cost. That was the first bet. We are sticking to that.
Then for the gimbal, in this system we had since the beginning a vision that we could have something relatively agnostic that could handle various types of connectivity.
After three years, what we learned is for GEO it’s not a big issue, it works with any type of GEO. For MEO satellites, it could be a good solution, so long as you stay [connected] when you have to swap from one satellite to another.
But this type of technology is not valid when you have to track a constellation and have beam hopping every five minutes. Then you totally lose the essence of what the NGSO constellation is offering, low-latency and so on. For this type of business, an AESA antenna where you could have multi-beam forming and one beam having to look at a LEO satellite realize at the same time another one is opening the link with this new satellite at the horizon, it’s absolutely mandatory.
Amy Svitak is a contributing editor of Space Intel Report.