5G’s Rollout Rattled Hundreds of Pilots
spectrum.ieee.org
spectrum.ieee.org
The 5G band that was auctioned off is only one step next to the band used by radar altimeters - essentially a piece of equipment on aircraft that gives it its precise height above the ground.
Aircraft can and do have barometric (pressure corrected) and GPS altimeters as well (which are usually giving you your height above sea level - and knowing the height of the terrain you can work out your height above ground), all of which feed into the various systems allowing them to have an accurate height at any given point during the flight.
This is most needed when flying “precision 3D approaches”, usually in bad weather when visual contact with the ground may not be possible until the wheels actually touch the ground. Autopilot systems that land the aircraft fly these “approaches”, and thus rely greatly on their height, speed and distance.
These radar altimeters were mostly designed and certified back in the 80’s (or even earlier!) - and it’s a lot of work to have any equipment certified - especially today. Hardware + software back then was nowhere near as good as it is today, yet many of these altimeters still being produced today would be to the original spec from a few decades ago - which had a much wider tolerance or sensitivity to the frequencies, so when you have a frequency only a few steps away - a slightly older / worn equipment with perhaps not the best shielding will definitely pick up some surges from frequencies next to it.
Blame the FAA for their archaic and slow, near impossible process to get equipment certified. Blame the Aircraft / Equipment manufacturers for not updating their equipment with the times, but that’s on the FAA for making this process difficult- see 737 certification as an example, leading to the max accidents.
The certification process being "difficult" isn't why the 737 Max was unsafe. The FAA allowed Boeing to self-certify its safety, and voices inside the company were ignored or silence. They literally created a system to save a buck for their clients then outsourced development of that system to save another buck, then rubber-stamped it themselves.
Blaming this on too much or too difficult regulation is farcical.
PS - I also find the entire thrust of this argument contradictory: "Equipment built in the 80s wasn't high enough quality for the new 5G operating environment, so we need to weaken 'difficult' regulation today to make producing lower quality equipment cheaper, tomorrow be damned."
They added the MCAS because they didn’t want the Max to be a different type, which meant starting the certification process from scratch. They were able to self certify precisely because it was a 737 type.
Yes - there were many issues that lead to the accidents and you can’t really point to just the one thing; but ultimately it did boil down to a culture at the company created by the environment it operates under.
I’m not defending Boeing here - if anything, I’m more upset at them as they had the engineering will and talent, the political and financial backing to make meaningful changes to the regulatory environment and make aviation safer as a manufacturer - but they chose greed.
Lastly, that’s nowhere near what I said but I can see how you can get to that conclusion. I’m an aviation enthusiast and a Pilot (feel free to go through my post history) - and it’s not a question of weakening regulation; it’s a matter of streamlining ways to get new technology certified faster - it only took 3 decades for the FAA to give the STC for unleaded gasoline etc etc.
https://seekingalpha.com/article/2765285-the-battle-on-the-n...
The shoddy software primarily responsible for the Boeing 737MAX disasters was introduced so that Boeing could market the planes to airlines as a standard 737 that required no expensive pilot training programs, as far as I can tell. The nature of the FAA's regulatory process seems mostly irrelevant here, unless rushing poorly designed products with huge risk profiles to market to help Boeing improve sales is something the FAA should promote... note also that Boeing lost around $30 billion in market value over the MAX debacle, which could have been avoided by FAA being more strict, even if that mean fewer plane orders.
Though, adding the MCAS, which in certain circumstances overrides the pilots instructions, should most certainly have triggered new training for the pilots for those situations, at the very least.
Nonsense. Boeing regularly does that.
> They added the MCAS because they didn’t want the Max to be a different type, which meant starting the certification process from scratch.
Also nonsense.
They added the MCAS because they didn’t want the Max to certify as anything but a different type because the entire point of the Max program was for customers who exclusively fly 737s (like Southwest) to be able to fly a more efficient plane without needing to recertify pilots.
Unlike Airbus, Boeing does not have unified profiles and accelerated cross-type training[0], so changing type is close to a full certification for crews, which translates to the crews being grounded for several weeks / a few months.
[0] IIRC they have something along those lines for some planes released close to one another, but Airbus has it across most of the range, with the exception of the older types and the A220 which is not an Airbus plane per-se (it's a rebadging of the Bombardier CSeries)
We are not limited to talking about 'weakening' or 'strengthening' regulations. We can talk about making saner rules, too.
No, the problem is that the altimeters didn't meet spec.
These altimeters are NOT supposed to be susceptible to these frequencies. However, people cut corners decades ago and now it's being found out.
The FAA should have told the airlines "You have N years to get these replaced. Start now." But they didn't--because they're in the pocket of the big aerospace companies.
Well, so the FCC rolled it out. Now the FAA and the aerospace companies have to deal with it.
The FAA's approach is reasonable because safety-critical equipment is, well, critical to safety. The FCC got rolled by the 5G hype train and failed to do its job - i.e. allocate frequencies to avoid unintended interference.
That's insane. No properly operating equipment needs that sort of gap, and certainly not radar altimeters built with the proper selectivity.
This is a 100% solved problem, and a 100% failure on the part of the equipment manufacturers, and whatever certification procedures exist for them.
It simply is not that hard to test this, certify this, etc.
220MHz is sufficient if your altimeter radar has the proper selectivity.
ground radars are so selective that you can have two radar devices operating at the same frequency aimed at each other without interference.
How do two devices on the same frequency, pointing at each other, not interfere?
additionally, the radars sweep through their bands rather than broadcast on all of it at once, so the chance of them looking for the frequency that the other radar is sending at the exact moment it is being sent is very small, and when it does happen it is easily ignored.
the sweeping is what provides the ability to easily detect movement and change in distance from radar to target.
this is all based on a book I read years ago, so I apologize if any of it is not exactly on the dot accurate.
A lot of people don't seem to grasp the magnitude of the spectrum lost to this snafu. At today's realizable spectral efficiencies, that is 2.5 TB/s of usable data throughput lost to the 400MHz guard band alone. It's also ~$130B worth of revenue that the FCC could have had if they could actually sell it.
The FAA is an executive administration ultimately organized under Dept. of Transportation. The FCC is an independent executive agency overseen, ultimately, by Congress.
As a result, neither org has unilateral authority to tell the other to go pound sand, nor do their up-org leadership (ultimately, the President v. Congress).
Their inability to find a resolution that didn't leave American pilots floundering is reflective of the inability of the Legislative and Executive branches to work together in modern federal governance, in general. Expect more issues like this in the future.
You don't put hurricane windows on buildings that have no hurricane risk, and you don't put tight bandpass filters on receivers in the satellite bands.
When hurricane patterns or satellite bands change, you get problems though.
Aviation is an industry where new products absolutely should not be fast tracked into deployment. The rules and regulations that are in place today were written in blood.
If anything it's the weakening of these regulations (i.e. manufacturer self-certification) that enabled the MAX disasters.
Your comment is a complete mischaracterization of the MAX incidents beyond what even Boeing public relations was willing to do.
The single point of failure was a bad implementation detail of a bad feature, and that feature mostly existed because of how all-or-nothing the "aircraft type" training is.
https://www.aopa.org/news-and-media/all-news/2019/september/...
The FCC only cared about selling the spectrum.
The FAA only cared about minimizing the chance of any accidents.
Neither was able to look at the science, come to a compromise, and then implement it in a joint way. It took the carriers, who had just paid $80b (from the article) to inject sanity into the approach via voluntary buffer zones.
Turns out, single purpose regulatory agencies aren't great at considering matters outside their area of focus...
It could. Did anyone bring it to a court?
This seems like a case of the FAA's "never change a running system" running into a world that is changing. A world that expects radio equipment to be better behaved than the stuff we built in the 80s.
And given their track record of success and reality, I partially understand. The only way to avoid unknown unknowns is to maximally simplify a system, and that starts with not introducing components uncharacterized by decades of real world use.
Obvious parallel in tech: database systems.
The problem is that aircraft inhabit the same world as the rest of us, so sometimes their environment does change from things outside FAA control. And at that point, delaying technology updates for nearly half a century becomes a major liability.
I think this is a good thing.
See for example the safety achieved by modern technologies in modern light aircraft designs, and the lack of certification of most of these because of the expense of complying with FAA certification requirements. So the certified fleet has to use existing certified designs from the 60s which have not progressed as much in safety.
The root cause of this problem is, to an extent, a problem of agency silos. The FCC exercised its responsibility for controlling radio interference almost entirely by regulating what equipment can transmit, but radar altimeters are now giving false readings because they are overly sensitive to signals they receive outside of their allotted bands. While this may formally be part of the FCC's responsibility, it is clearly something the FAA could have been more concerned about.
If we only cared about safety, zero planes would be in the air right now. But sometimes usefulness outweighs safety.
And the money-grubber argument could easily go both ways. The airlines bought radar altimeters that were out of spec. They could buy radar altimeters that are in spec and just as safe, but that would cost money. You could just as easily spin that as the FAA only caring about money.
The only way to achieve total aircraft safety is to have no aircraft. If the FAA literally cared about nothing other than safety they would have everything grounded, so clearly they do to some extent care about other things.
Plausibly, a well-designed government would be explicit about the tradeoffs between safety and convenience and cost and so forth. In practice, that might be politically catastrophic. What usually happens, I think, is something like this: the official line is "keep risk down to a negligible level", where "negligible" is usually not quantified, and cost/inconvenience is largely ignored until everyone agrees that the risk has been made negligible, but below that threshold everyone's free to minimize cost/inconvenience.
(This might actually end up being a pretty good strategy. Let's measure cost and convenience in dollars, because trading those off against one another isn't so problematic; then maybe a reasonable approximation to what we care about is that 1. each death is like burning $X where X is probably some number of millions, and 2. having any major accident in a given year is like burning $Y where Y is many billions, because if there's a major accident then the hit to public confidence means (a) big loss to the airline industry, (b) more people driving instead of flying, which means more deaths because flying is much safer, and wasting a lot of their time, and probably also more environmental damage. In that case, maybe the effect of #2 is actually not so very different from "big penalty if the risk goes above a negligible level". But it seems like we'd get better decisions overall by being more explicit about what we care about and how much.)
An approach of joint utility, taking into account time, money and risk usually works better.
Give me a break.
Everything is a trade-off, and actually what we usually aim to do as society is to manage acceptable and balanced levels of risk rather than use safety as the single-lens to make decision. As an example, the design of cars would be very different if they were designed only with safety in mind, and if automakers pursued this goal ignoring factors such as speed, comfort and price point (or indeed, competitive pressures).
It's also worth noting that the "efficiency" in my post actually came from the stated goals of the FAA.
Retrofitting band pass filters on existing aircraft radar altimeters isn't a trivial task - avionics changes to aircraft are a multi-year project since extensive testing is done on anything electrical installed on an aicraft.
Additionally, some 5G infrastructure equipment may need additional band pass filters installed to prevent spurious transmissions outside of their assigned frequency ranges in areas near airports.
If you aren't familiar with a bandpass filter, it basically filters all RF outside of a certain frequency range. Sensitive receivers can naturally be overwhelmed by RF outside of the receiver's frequency range, and preventing that RF getting to the receiver in the first place is the role of a band pass filter.
In this case we can see it’s not the usual suspect (over/efficiency mindset) but maybe there is something much uglier lurking underneath? Affecting both companies and government bodies even - which should feel less “budget” pressure to be overly efficient in everything?
I don’t know what factors those are. Though they could still be that plus something else (more transparently cultural)?
Businesses are increasing consolidated and have market pressure to show not just profits but increasing profits year over year.
Government agencies are affected by a combination of regulatory capture and a half century of austerity
The trick is to make it look more complicated than that, so no one follows the money. If somebody does, call them names. If calling them a Luddite conspiracy theorist doesn't work because you used to employ them as an expert and they have a long track record, just say that they're bolstering or emboldening or platforming Luddite conspiracy theorists (who might also be racist-adjacent China-deniers.)
You buy a media in order to use it.
5G has not appreciably made “flying more unsafe.” If anything, this is a good push to start updating altimeters to spec (installing bandpass filters, better testing) which actually will make flying more safe because currently these altimeters are susceptible to out of band frequencies.
Every damn time.
Sometimes I'm able to mess with the network settings to turn off 5G and go back to 4G, and then my phone works fine again.
https://www.salon.com/2022/03/14/how-5g-could-send-weather-f...
The central issue here is greed. The bands used for science (like 23.8 GHz) are critical, but not enough concern was given to them, so neighboring bands were auctioned off without proper consideration of consequences from interference, due to lobbying pressure from telecommunications companies. From the article:
While the FCC's 2019 auction of the 24 GHz spectrum band generated $2 billion in revenue for the Department of the Treasury, the costs from severe weather could be much greater.
This is the type of unintended consequence that comes with privatization of government services. What looks like a cost savings in the short term ends up being a cost increase in the long term. See also: the commons, natural monopoly.
Is there any evidence this is the case? This sort of thing is a lot like property zoning. If a steel mill starts up next door to your residential property in what used to be a park then there is nothing stopping you from adding more sound proofing to your walls to make it so you can sleep. But should you have to do this and can you do this at all? In this case the neighbouring band was originally used for satellite downlinks. So barely detectable signals with large dishes required. Radar altimeters in aircraft actually used to cause interference to the satellite downlinks when the aircraft flew over.
The fact is that you need to have sufficient physical separation between services in adjacent bands. In this case, you can't expect things to work if an aircraft flies directly over a 4 GHz 5G transmitter at a low altitude. This aspect of adjacent band interference is an important part of spectrum management. You can't just wave the issue away...
It's more like the family who is first to move into a neighborhood under construction, complaining that their kids can no longer play in the empty lots because houses are being built there. And it's not even the empty lot next to them, but one lot over.
This sort of thing is why allocating TV frequencies is and was complicated. That is because someone might live right next door to a TV transmitter and they might be trying to watch a weak signal on an adjacent channel. You usually have to allocate some dead channels for the area to prevent receivers from being overloaded. So in that case would it be fair to say that receiver was using channels it had no right to when set to the weaker signal? After all, the transmitter was there before the receiver was...
Note that other countries have managed to deal with this routine and common issue with respect to the introduction of 4 GHz 5G. There is something weird going on in the USA...
Then what is the required passband attenuation at 220 MHz away from the centre frequency as per the spec? What altimeters are not meeting this spec?
I find it weird that people are claiming that the altimeters are defective here. That would be a big deal if true.
If we go this way we would have to defund highway patrols because it is waste of time (traffic rules already specify how to drive safely). You don't need IRS (because law says how to pay taxes).
Also try to understand those pilots did not manufacture the equipment they are flying. They bought equipment that was certified to be airworthy or are flying for somebody who owns the equipment that was certified to be airworthy.
This is like standard practice for shit that is of trivial importance to humanity in a newish engineering field. So I don’t for the life of me understand the logic of being more flippant when it comes to human lives in a more established engineering field.
I see 3 solutions: 1: shut down 5G. 2: Ground airplanes with defective equipment 3: Make the airplane companies pay for the spectrum that their defective equipment needs to operate.
Option 2 would satisfy flying safely and not put externalities on companies that had no fault in this. 3 also would but would not be very efficient.
No they didn't. The equipment worked just fine until people started broadcasting on a nearby frequency.
Like if I put a jammer outside your living room window and your WiFi stopped working, would that be because you bought 'defective equipment'? After all, your router doesn't have the latest 'anti-jamming' error correcting codes built in.
A jammer is a malicious, unlicensed product specifically designed to prevent your Wifi from working.
Regardless of what the specifications say, you can't rollout new technology without taking the real world parameters into account, and these were well-known parameters before anything rolled out.
From their perspective it worked for decades and now someone decide to jam it.
From engineering perspective, yes, they are defective and don't have enough filtering on the input to stay within its band
I know it's more complicated and costlier than that but since lives might potentially be in danger and since it's too late to change the 5G bands, it seems to be the only viable option.
If it it indeed causes, C band can simply be banned near runways and say, be allowed inside the airport in low power settings if it doesn't cause interference with outside planes.
5G base stations have compliant out-of-band emissions so the issue is not "because of 5G" but rather "because radio-altimeters gather energy from adjacent bands" (something we call "blocking"). Not all radio-altimeters are the same, but for some of them there is very little filtering and the only thing that protected them from good old 4G was that the antenna behaves somewhat as a filter and 2.6 GHz was far enough from 4.2 GHz... It may sound silly for a safety-critical equipment, but we really have to think about the really long lifetime of those equipments and think with the context of the time when they were designed/deployed (e.g. if there is only space-to-earth transmission in adjacent bands, why would they put filters that would add insertion loss and cost ? In the 80's there was no 4G/5G in sight...). Probably the aeronautical authorities also failed to inform the relevant issue in due time within the spectrum regulation process, so this has not been known or studied properly at ITU/CEPT/etc in the context of those frequency bands close to 4.2 GHz. It also seems that the radio-altimeters vendors keep a lot of secrecy on the technical details of their equipments (a part of the internal culture which also comes from the dual civil/military use of those systems) so even people who represent civil aviation authorities or aircraft manufacturers such as boeing/airbus in spectrum regulation authorities do not always have full access to the relevant information.
Now the situation is what it is, and the short-term solution is to have some kind of 5G exclusion zones around airports, and of course operators are not happy with this because they paid huge prices for 5G licences and now they discover that they have new/additional constraints around airports... Long-term solution is also an issue, because it is not that simple to put filters on existing equipments (if they change the behavior, the whole computer need to be re-calibrated, and pilots trained, and the whole thing re-certified, which is also not as simple of doing it once and certifying all aircrafts from the same model but really must be done one-by-one in a custom way for every aircraft). I understood that new specs for radio-altimeters will soon be out so we can hope that the situation improves soon as the retrofit will be able to start (even though it will take years), but there is a possibility that even new radio-altimeters may still continue to be vulnerable to blocking in 4-4.2 GHz in order to maintain their required accuracy (I'm no expert in radars but I trust those people who told me). So the issue will be solved with regards to 5G in 3.4-3.8 GHz as deployed in Europe, but maybe not above 4 GHz (which is one reason why it may be better to restrict to low-power/verticals in the future uses in those bands).
Anyway, this is a complex topic as you see...
This is just pure speculation.
From the article it sounds like anyone can go to an airport and simply send some garbage signals towards an approaching aircraft to make it deploy air breaks and crash if the pilot is not quick to react.
When I think about it it seams like a common thing in aviation tech. ads-b has similar issues, it is completely unauthenticicated/unsigned there is 100% trust.
Anyone can pretend to be anyone else.
Relevant discussion:
* https://www.intelsat.com/wp-content/uploads/2021/02/intelsat...
Why? It is not like there is serious privacy violation in telling that altimeter broke, and this makes the data much less reliable. Not saying the data is fake for sure, but I have came across enough haters of 5g, and I am pretty sure someone could post fake entries just to scare people.
The goal is data collection to assess and prevent incidents. NASA's intake on the reports are not anonymous, but their public reports are. NASA functions as the filter between pilots and the FAA/NTSB.
If pilots are not afraid to report incidents, they will report them more often. This in turn means that hidden issues are more easily discovered, which in turn could save lives.
This is primarily directed at lightning.
Just a guess though.
Thus you can pretty well observe a lot of what really is 4G as 5G.