We’ve heard about 5G networks for mobile phones for years. First, they were going to be super fast and amazing, then they were said to cause COVID, and now they’re predicted to bring the airline industry to its knees. That’s quite the resume for a little ole’ cell phone network. But here we are and the issue has exploded. What exactly happened? I’m going to do my best to break this down.
How C-Band 5G Came to Be
The term “5G” is just the next generation faster network that goes above and beyond 4G LTE. The mobile phone companies are all racing to roll out their networks as quickly as they can so they can crow about having the fastest, the most, the best, etc. The problem is finding enough spectrum in the right frequency to be able to serve the needs of the customer base of all the providers.
For years, there was a debate on how exactly to free up more spectrum, and in the waning days of the Trump presidency, it was finally auctioned off. The so-called “C-Band” mid-range frequencies were used by satellite operators, but the Federal Communications Commission (FCC) decided to kick them out and reallocate to the highest bidders. (They haven’t all been kicked out yet, but they will move up the ladder by the end of 2023, I believe.) In the end, AT&T, Verizon, and T-Mobile paid a combined $81 billion to get their hands on this chunk of spectrum. T-Mobile only picked up a small piece of it to supplement the network, so this is really an AT&T and Verizon story.
What those two providers got was in the 3.7 to 3.98 GHz range, a mid-band frequency that is good for mobile phones, because it travels long distances and goes through walls. But it’s also pretty close to the 4.2 and 4.4 GHz range that is used by radio altimeters on airplanes. There’s been concern about whether this explosion of use in the 3.7 to 3.98 GHz range near airports might impact the functioning of those radio altimeters.
The Mighty Radio Altimeter
In simple terms, a radio altimeter uses radio waves to figure out how high your airplane is off the ground. There are other kinds of altimeters that use air pressure to determine altitude, but the radio altimeter is important, because it ties into various systems on the airplane. Most notably, it is used to determine altitude while on an autoland approach in bad weather. It’s also an integral part of the heads-up display system in some airplanes, and it is responsible for triggering reverse thrust and spoilers upon landing in some fleets.
How often does an airplane need to use those things? Well, if the visibility is low enough, you need that radio altimeter to get you down on the runway safely. Ever seen Die Hard 2? You know when they change sea level so that Windsor Air crashes on the runway, thinking it’s higher than it really is? Well, that’s all absurd, BUT the point is kind of the same. If the airplane thinks its at a different height than it is, bad things can happen.
Further, the heads-up display is a great safety tool, and if you can’t use that, then it does degrade safety over what’s possible. And a delayed deployment of thrust reversers and spoilers won’t be a big deal unless you’re on a short and/or wet runway. Then it could be bad news.
The bigger issue here is that the Federal Aviation Administration (FAA) stepped in and declared that certain aircraft couldn’t be used at certain airports if the C-Band was flipped on. Is there really an issue? We don’t actually know for sure, but the risk is enough that the FAA decided it was worth stepping in… all because of that little radio altimeter.
With this reality that about half the US fleet wouldn’t be able to fly as normal, things got ugly very fast. It wasn’t just US airlines that panicked. Emirates, ANA, JAL, and others said they would stop some flights. Other airlines furiously tried to swap airplanes out for US flights at the last minute, using ones that had radio altimeters that had been cleared. This rush at the last minute was mind-boggling. How did we get to this point where it all ran up against a deadline? Shouldn’t this have been resolved before?
Shouldn’t This Have Been Resolved Before?
The answer to that is yes, but when you get several competitors and political bodies working against each other, things don’t go very smoothly. On the one hand, you had AT&T and Verizon not wanting to share what the other was doing. Once the C-Band auction was done, as I understand it, the wireless carriers could do pretty much whatever they wanted with it around airports. Then you have the political fight between the FCC and FAA.
The FCC sold off the C-Band spectrum, and many have wondered… why didn’t the FAA stop the FCC? Well, it couldn’t straight-up stop the FCC, but it did voice its displeasure. It tried to get the auction delayed when it happened at the end of 2020, but the FCC brushed it off, saying everything would be fine. The military did the same since it happens to fly a whole lot of airplanes itself. The RTCA put a study out in 2020 saying in part that “the results presented in this report reveal a major risk that 5G telecommunications systems in the 3.7–3.98GHz band will cause harmful interference to radar altimeters on all types of civil aircraft….”
Apparently all this fell on deaf ears, but that doesn’t mean the FAA should get off unscathed here. Maybe if it hadn’t waited so long to ban certain airplanes from flying after 5G deployment, then people would have listened more. Then again, the FCC under Ajit Pai seemed hell-bent on making this happen regardless. And now, here we are.
But, But, France
I know what you’re wondering… why can they roll this out in Europe just fine, but here it’s a massive problem? The devil is in the details, and the reality is that these may both be called 5G rollouts, but they are not the same thing.
France is one country that seems to be used as the posterchild for why aircraft can fly just fine with 5G operating, but what happened in France is very different from what’s happening in the US. In fact, I’m going to just steal this image from the FAA’s 5G page that has recently gone live.
The French antennas have permanent safeguards in airport buffer zones that provide more protection than the US ones. Further, the French antennas near airports have to be tilted downward to reduce interference, and the French antennas have far less power. Not mentioned here but also notable is that in Europe, the C-Band spectrum is in the 3.4 to 3.8 GHz range, so it’s further away from the range that radio altimeters use.
In the US, it’s just pure madness. Either regulatory authorities completely failed here, or the structure of the regulators just isn’t set up to properly handle this situation. Either way, it’s bad, and now we’re up against this deadline where everybody is pissed off.
The Solution
What’s the solution? There are so many. The most obvious is what seems to be happening now. AT&T and Verizon have — not because of regulation, mind you, but from pressure being exerted on them — limited their rollout near airports. AT&T sounds particularly dickish with this statement it gave to NBC.
We are frustrated by the FAA’s inability to do what nearly 40 countries have done, which is to safely deploy 5G technology without disrupting aviation services, and we urge it do so in a timely manner,
In the long run, presumably there will be more testing to prove which radio altimeters can function properly and which can’t. And eventually those that can’t, if there are any, will be phased out and replaced with something that doesn’t cause issues.
For now, however, just look for a lot of finger-pointing and confusion. It’s the American way.
23 comments on “An Attempt at Explaining This 5G Mess”
Altimeters that aren’t masking out of band signals should be retrofitted or replaced. That will take some time, something so mission critical should have been addressed years ago the negligence here is shocking, but should now be prioritized.
This has been my question the whole time. Why didn’t airlines upgrade the radio altimeters? Given the potential safety implications, it seems like it would be a prudent change given how crowded the airwaves are becoming. Even if the US implemented the same 5G requirements as France, it seems like there’s still some risk of interference.
I suppose it all comes comes down to money. I suspect airlines would argue that wireless companies are encroaching on their spectrum and thus they should pay for these changes.
Regardless, it seems like a colossal failure by everyone involved.
I’m no fan of the way the telecoms and the FCC conduct business, but to be fair, it’s pretty sloppy engineering if something designed to receive signals from 4.2 – 4.4GHz can fail to work properly if it receives a signal more than 200MHz away from what it’s supposed to be looking for.
The FCC has been selling off frequencies for years. They have sold most of the frequencies that were used by the wireless microphone industry forcing all users of wireless microphone equipment to replace perfectly good equipment on their own dime including Houses of Worship. The FCC gave about as much concern for the confusion they caused then as they are with this situation.
Well, again, it really wasn’t “perfectly good equipment” was it, if that equipment was using frequencies that weren’t allocated to that use in the first place. That’s not the end user’s fault of course, whoever they may be, but their ire should be directed at the equipment manufacturer in this case.
DP
It does not have to be “sloppy engineering” for a radio trying to receive on one frequency to pick up a strong signal many Mhz away.
I am an amateur radio operator and I can tell you from personal experience that close relatively strong signals can easily swamp the input of a sensitive receiver no matter how much filtering is added.
Per the Honeywell paper on the topic (https://ecfsapi.fcc.gov/file/7021340930.pdf):
“By the physics of their design FM?CW radar altimeters have exceptionally sensitive receivers with
minimum detection thresholds between ?110 dBm and ?120 dBm”
For your reference DP, here is the definition of dBm:
A 1-mW (one one thousandth of a watt) signal has a level of 0 dBm. Signals weaker than 1 mW have negative dBm values; signals stronger than 1 mW have positive dBm values.
So the 1500 watt transmitted signal that AT&T and Verizon want to use, could easily cause distortion or front end overload at the very sensitive radar altimeter receiver..
This is probably the most straight forward and accurate description of the issue that I’ve read; thanks, Cranky! The Die Hard 2 reference really brought it home.
I dug into the RTCA study a few nights ago and it’s kind of appalling that we got to this point. Complete negligence on the part of the FCC. As Tim Clark at EK stated, “This is one of the most delinquent, utterly irresponsible issues, subjects – call it what you like – I’ve seen in my aviation career.” And he’s right. The FAA raised their concerns as did the airlines and the FCC bowed to it. Just like they did when it came to Net Neutrality.
You undersell the importance of the radar altimeter a bit. A *lot* of the airplane’s logic flows through the radar altimeter on every flight, as the safety envelope and automation behaviors are usually tied to altitude. For example, he availability of the go-around mode and how it is engaged change a couple times every approach depending on radar altitude. Landing in bad weather might be the most critical example, however, as an erroneous indication of low altitude on some planes can trigger the autothrottle to retard to idle.
Fault at some point must come back to aviation authorities who have for decades allowed radar altimeters that do not stay within their allotted frequency spectrum. Aviation was never allotted < 4 GHz and has no business being there. Those cell towers will pump out huge amounts of energy on adjacent frequency bands without any interference whatsoever because they have modern radio equipment, an airplane listening 200 MHz from allotted frequencies is just absurd. Frequency filtering isn't particularly advanced tech.
Fault must also belong to the FAA, who has had years to evaluate specific radio altimeter installations for performance, and has failed to do so for most planes. Last week they had cleared 45% of airliners, I think, and this week 62%. They seem to have been operating on the belief that by doing nothing they'd force the FCC to blink, and the FCC did not, so FAA is badly unprepared.
It's ridiculous, too, that we still use the 1960s standard of multiple redundant sensors for safety instead of having any sort of sensor validation. There are a dozen different ways for a plane to locate itself in three-dimensional space that could be used to validate an RA to the accuracy needed for everything but perhaps the flare and touchdown, yet none are used. It's all-or-nothing on the Captain's RA. This is a good lesson for everyone shouting about the inevitability of single-pilot airliners with the current generation of planes – this entire mess is because no flight computer is yet resilient against a faulty radar altimeter or any other faulty sensor and pilots are, as they will be, the only backstop.
Eric C – Fair point. I didn’t want to get into the full ins and outs, but then again, I probably don’t know the full ins and outs. This is a really helpful comment, much appreciated!
Cranky:
The below document should give you many of “in’s and out”s”.
It may be a little too technical in spots but it does a pretty good job of explaining the issues with 5G interference with and aircraft’s Radar Altimeter.
https://ecfsapi.fcc.gov/file/7021340930.pdf
That author might be brilliant with radio frequency engineering but definitely does not understand flying particularly well. Autopilots do not deploy flaps and gear, that is not a possible consequence of an RA spoofing event. EGPWS has a high-res terrain database that could be used to derive height from altitude, but that’s not something anyone has tried yet.
Parts of it are over my head, but I still can’t grasp why a radar altimeter is bothered by frequencies 200 MHz away from its operating range. From reading what some radio engineers say that sounds like a huuuuge amount of territory for modern systems. Supposedly frequency filtering is cheap and easy these days, too. I wonder if these planes the FAA are approving are ones that have proper filtering built in? If so, why the kerfuffle? If not, how are they being approved?
From a communications engineering perspective (which tends to put most of the blame on the FAA and AVSI) see https://wetmachine.com/tales-of-the-sausage-factory/what-the-eff-faa-my-insanely-long-field-guide-to-the-faa-fcc-5g-c-band-fight/
Both the George W. Bush and Obama Administrations invested a fair amount of White House clout in herding the cats and making the system work. Agencies don’t have authority over each other, but the executive branch reports to the White House. One specific federal agency, the NTIA, is designated by statute as the official Executive Branch interface with the FCC (with a couple of exceptions, such as the Small Business Administration Office of Advocacy). The Obama folks in particular invested time and effort from the Office of Science and Technology Policy (OSTP) and the President’s Council of Advisors on Science and Technology (S.H.I.E.L.D.) to develop a framework (The PCAST Report). Trump, by contrast, repealed the PCAST framework and never developed any replacement. Trump blessed then-Chairman Ajit Pai’s “5G Fast” plan, told Pai to win the 5G race, and then moved on. The Executive Branch agencies responded by ignoring the required federal process, trashing the FCC in the press, and trying to run proxy wars with against the Commerce Committee with competing committees in Congress (notably the Armed Services committees and the House Transportation Committee). Things got so bad that the Government Accountability Office gave testimony and issued a report which explained (politely) just how screwed up things have gotten.
Geoff – Oooh, I like this perspective. I’m going to post in 3 Links tomorrow so more people see it. Thank you.
Thank you for explaining this situation in layman’s terms. Another aspect is the extent to which this can potentially harm operations. Consider DFW airport, for example. There is a litany of instrument approaches there that are no longer usable. Basically, any approach that requires a cloud ceiling below 200 feet and visibility less than 1/2 statute mile is no longer available. In addition, the Next Generation RNAV/RNP approaches, which move air traffic more quickly in low ceiling and low visibility conditions, are also banned. Basically, we have gone from modern instrument approach conditions (Category 2, Category 3, and RNAV/RNP) back 50 years to Category 1 instrument approaches, at the affected airports. This will certainly lead to more cancellations, more delays due to airborne holding (waiting for the weather to improve), reduced ATC efficiency (due to “lost” RNP/RNAV availability), increased carbon footprints, increased costs to the airlines (eventually leading to higher fares), and greater inconvenience to the traveler.
The manufacturers, the FAA, the airlines, the airports, and the military have spent decades developing the technology, proving its efficacy, improving airport approach and runway lighting, developing auto-land capabilities, delivering Heads-Up-Displays (HUD) that are so accurate that they are flown WITHOUT an autopilot down to fifty feet above the ground. ALL of this capability is now erased at the affected airports. Places that are fog-prone (SEA, SMF, SFO, etc) will now be taking huge delays and cancellations on low visibility days. When you start taking an entire hub out of action for hours at a time, it WILL impact an airline’s entire system. As a result, it is incumbent upon the FAA and the A4A to prevail upon the telecoms to modify their hardware similar to the European Model. Of course, a few inconvenienced Congressmen and Senators might be the industry’s best advocates.
I thought this article also does a good job of explaining the issue. Overall, it looks like the FAA is the most culpable party in this fiasco. https://www.techdirt.com/articles/20220118/08262548305/airline-ceos-freak-out-over-5g-despite-limited-evidence-real-world-harm.shtml
Outstanding presentation of the issues, CF
A couple key points stand out.
Deployment of 5G around airports is NOT the same as other airports; the FCC didn’t even understand what other countries were doing until well into the process. AT&T and VZ could have solved the problem if they had voluntarily configured 5G near airports as other countries have done.
Aviation has been using the frequencies for years which C band 5G now wants to use and the FCC auctioned those off. Previous satellite use is far different from cell towers. It is not aviation’s responsibility to demonstrate safety of 5G to their aircraft when their equipment was built based on then-established use of those bandwidths and they also shouldn’t foot the bill for retrofitting their equipment – which is a root piece of this discussion even if it is not being said.
The structure of the US government has no mechanism to resolve inter-agency conflicts which is why aviation’s concerns went unheeded. That structure has to change whether it involves aviation or any other two agencies.
It is unacceptable for an industry to be producing its own research and oversight (either pro or con 5G) and showing it to government for approval whether the topic is 5G deployment, Boeing’s MAX issues, or Pfizer’s research about when covid vaccines are “needed” which, has translated into billions of dollars of their own sales. Other countries have their own technical experts which are independent from industry and the US simply must spend the money to do the same – and put appropriate fences between industry and government experts.
Ultimately, AT&T and Verizon buckled because no one wants to be blamed for hundreds of airline delays or cancellations (United has already tried) or, worse yet, an aircraft incident or accident. When airlines have to resort to that level of manipulation of the media in order to get the problem resolved, the system is broken and everyone pays the price.
Btw, I have been a happy TMobile 5G user for a year. It is a quantum step up in cell service – but it can’t happen at the expense of people’s lives or even the risk of it.
CF – While I have to congratulate you and your team for an excellent briefing on the issues here, I also have to thank those commentators who have weighed in on the topic! It is so helpful when your readers CONTRIBUTE to the discussion rather than be reduced to unpleasant banter (to put it mildly!) Many thanks to all!
I just found this youtube video from a 777 captain explaining the issue with 5G and radio altimeters.
I think he does a great job of explaining the issue from a pilot’s view.
It is 28 minutes worth watching.
As an electrical engineer I have to say that failing to filter out of band transmissions is unacceptable. If I purchased a radio or TV that didn’t work because it couldn’t separate out two adjacent stations or stopped working because the FCC reallocated an unused adjacent channel one would be hard pressed to blame anyone other than the receiver manufacturer. Years ago when the possibility these new frequencies would come into use appeared remote perhaps it was a reasonable cost/weight saving decision but the altimeter manufacturers should have always been prepared to properly filter out frequencies not assigned to them when the situation changed and begun doing so as soon as it became clear filtering would be necessary for their equipment to continue to function properly.
Blaming this on the FCC makes no sense given that there is no evidence, nor have the altimeter manufacturers alleged, they can’t co-exist with the new use (they only state their current equipment without filtering may have problems). That the FAA failed to require changes isn’t an excuse either. The manufacturers knew they had made a decision to forgo filters and that the consequences of this would be problems when 5G came into use. At that point the airlines, FAA and the manufactures knew that they had a problem and it was their problem to solve. The only reason I can imagine they didn’t was that they didn’t want to pay the cost for their earlier decision to assume these adjacent frequencies wouldn’t come into use. So this is about nothing more than money…who is going to pay to upgrade or replace equipment that was based on the unreasonable assumption altimeter manufacturers could prohibit the use of frequencies well beyond that which was assigned to them.
BTW forcing cellphone companies to reduce signal strength or tip their antennas earthward reduces the coverage or the 5G signal and wouldn’t necessarily fix the problem since the radio altimeter would still be susceptible to out of band interference due to lack of filtering. This is clearly shown in the chart from the video. High frequencies, such as the are more like light than the low frequencies AM radio uses. They bounce off building and other structures in all kinds of way both interfering with and reinforcing the signal strength of the transmission so it’s not a good plan to try and do all kinds of stuff with the towers rather than fix what is clearly poor design with the altimeter receivers. It’s probably the only option while the altimeters are upgraded or replaced but is not in any way consistent with good engineering design. The airline technical people at the airlines should have been all over this demanding a retrofit solution from the instrumentations. Maybe they were and management didn’t want to eat the cost. It was the airlines and manufacturers that benefited from those adjacent frequencies not been used and therefore the savings that resulted from not filtering. But the free ride is over. They need to retrofit whatever upgrades are required (most likely an inline filter, which appear to be commercial available). It’s not like they haven’t known this was coming.
I would like to add some perspective regarding the scale of the portion of spectrum in question: 0.2 GHz is equal to 200 MHz.
Look at the FM dial (I know, that’s something an old guy would say but still…): the FM band starts at roughly 88 MHz and runs to 108 MHz. VHF TV starts in the 40’s and ends at 87.75 MHz. The separation between the C band used by 5G and radio altimeters actually dwarfs all of this. If you have a 150,000 watt grandfathered FM signal at, say 93.7 MHz, this thing doesn’t render the rest of the band unusable. Yes, on receivers with poor selectivity there can be bleed in right by the transmitter. But other FM signals co-exist happily in the same market as that super powered signal. Yes, in the old days you’d ‘see’ bleed in from a super strong nearby FM transmitter bleeding into an analog TV channel at times – but this was limited and these “shadow signals” occur at harmonics that are far closer to the main transmission frequency than a whopping 200 MHz.
So how is it that these frequencies, that are actually relatively far apart, can’t be separated out? Does the equipment on certain aircraft have such poor selectivity as to be interfered with by signals that are so off-frequency? Furthermore, the 5G transmissions don’t have the wattage of my theoretical super FM signal above. Yes, they have redundant transmitters, but none should have the oomph to overpower communications at a full 200 MHz above their spot in the spectrum.
Before I get jumped on: LOW VHF TV (channels 2-6) is the one that starts at 54 ends at 87.75 MHz. High VHF (7-13) is well above the FM band and goes from 174 MHz to 216 MHz. Meant to write that but forgot. There are literally nav beacons in the part of the band just above 108 MHz – that separation from FM is infinitely smaller than we’re talking about here.
Look at the chart. The radio altimeters mask (ie filter) goes well below 3.7 GHz. It’s not that they can’t filter, it’s that they didn’t and when it became apparent they would need to they failed to take action and now the representatives of the companies are shocked, shocked that transmissions way outside their allotted band might cause their equipment to fail to function properly.