Hacker News Reader

Comment by jrockway | Hacker News Reader

A lot of comments seem to be about spoofing. Remember that there are multiple humans in the loop -- experienced pilots and experienced air traffic controllers -- and multiple levels of redundancy. Pilots have access to barometric altitude, radar altitude, traditional VHF navaids, TCAS-issued resolution advisories (based on directional antennas, not GPS)... in addition to GPS. Air traffic controllers have primary radar and secondary radar in addition to ADS-B information.

You could probably cause a lot more trouble by buying an aviation radio off the shelf and just chatting on the frequency or issuing fake clearances if you want to be a mass murderer.

Ultimately the system comes down to many systems working together, not one stream of commands between two computers.

View on HN

> buying an aviation radio off the shelf and just chatting on the frequency or issuing fake clearances if you want to be a mass murderer

Incidentally, I am happily surprised that that never seems to happen. One could so easily create so much chaos at an airport... There is no authentication whatsoever.

But the risks and consequences are catastrophic. The Feds will put you away forever for even attempting it.

While true, you'd think state sponsored terrorism wouldn't care about sacrificing a few people.

What kind of state sponsored terrorism? Black ops programs? Air forces dropping bombs? Assassinations and decapitation strikes?

Or are we talking about the boogieman that is an excuse for all sorts of paranoid crackdowns, expenditures, and questionable legislature?

Who exactly are you afraid will sponsor radio jammers at airports, and why do you think they will do so?

actually, I'd think state sponsored terrorism would have the resources to get away with it.

Just send the voice data through a raspberry pi to a remote location and carry out the attack from there. You could send voice data through a TOR-like protocol or something to keep you anonymous, if that's important. You could maybe rig some explosives to go off if the FBI/FCC/FAA gain access to the attack point, to prevent them from recovering evidence.

Lot's of police/fire radio systems are similarly without authentication, so you could amplify the mayhem by doing the same thing to those frequencies.

With state sponsored terrorism, the risk isn't that the Feds will put the agent away forever, it's that the USG (and, incidentally, other particularly allowed countries) will identify that it is state sponsored terrorism and treat the attack as exactly what it is—simultaneously and act of war and a gross violation of the laws of war.

And most major airports have systems like RFEye or AIRPORT-SHIELD. You would be _immediately_ pinpointed with a remarkable level of accuracy from 3-6 sensors.

Especially near major metro areas, unintentional interference is a daily issue. These systems immediately triangulate sources of interference in a wide band. Or you have simpler systems like UMS100 that just do alerting and you send out an engineer with a wand to walk around for a bit to triangulate it.

These could be pretty easily defeated through a multitude of tactics. They're just being pitched as a stop gap solution against the low hanging fruit like morons with drones or a drunk person with an airband handheld.

Imagine an unlicensed drone emitting GPS radio signals. Just a small amount of sophistication and an attacker can do an incredible amount of harm.

If you live in Florida, you can go down to the pawn shop, buy a gun, drive over to the airport and start shooting at planes. Doesn't take much sophistication to cause a lot of harm.

At a plane. Then you get a SWAT response and terrorism charges - consequences are also a reason this doesn't happen on a mass scale.

With an omnidirectional antenna, sure. With a high-gain antenna, no. The attacker could entirely avoid the RFEye or AIRPORT-SHIELD. The attacker could pick a specific aircraft to pester, with no other aircraft able to listen.

Add some fancy voice synthesis, and it gets interesting. The attacker could clone the voice and style of the air traffic controller. The attacker could provide enough power to quiet out any other transmissions. To keep the air traffic controller incompletely aware, the attacker could provide interference (just to the tower) whenever the victim aircraft begins to transmit.

It happens sometimes [1]

It actually happened near me once when a local crazy guy got mad that planes were flying near his house (he bought a house right across the street from the municipal airport) so he found an airband radio and started yelling at them every time they came too close for his liking.

[1] https://www.youtube.com/watch?v=ZvA_-linhg8

Wiarton is pretty far from Toronto. But I dunno how many receiving/sending antennae they have for « Toronto Centre ».

Toronto Centre is in the top level of the ATC hierarchy and is a geographic area rather than specific to the city of Toronto. Airband AM only has a range of 50-200mi in flight so they have outlying stations, usually colocated with VOR stations IIRC.

He probably got lucky and had a station within 5-10 miles of his bathtub...

https://en.wikipedia.org/wiki/Toronto_Area_Control_Centre

I'm kind of wondering if there even needs to be. If you keyed up on frequency near a major airport as described in the GP, I'm pretty sure the FAA and FCC would be working together to track you down in short order. I could see the argument that it would be good to add authentication to the radios to complement the strict enforcement for a sort of belt-and-suspenders approach. However, I think a more pressing concern for aviation radio is that is is kept simpler, and thus with fewer failure modes, and thus more reliable Because when you need that radio in an emergency, that has to be the worst possible time to have to deal with authentication hiccups.

Could they? A moving target? Say the soundproofed back of some van.

I personally would volunteer to be on one of the teams in another van with directional antennas to triangulate your transmissions and shut down you ASAP.

The fact that no one has ever truly tried that sort of poisoning of radio calls is more due to the tower and aircraft having a very well defined set of calls, so that the likelihood of false transmissions being able to cause havoc is minimized already.

I agree with another commenter that aviation radio systems should be the simplest, most reliable systems possible, and have that layer of communication depend on proper radio call discipline and no more.

My question was of theoretical nature, i'm not up to mischief of that sort. Though i'm often thinking about the vulnerabilities of our technological society, and the (false) assumptions we have about that. Compare it with lock picking. Most are easily bumped, or not even bothering with that, the door kicked down, some window used instead, and so on. Now, when someone kicks your door down, that's getting noticed, sooner or later.

But what about cheap SBCs dropped somewhere with a battery and some custom periphery working as a relay? Of course the SUVs could find that. After the fact.

Pulling these kinds of attacks - especially undetected - requires a certain level of intellect and education that should also be unaffected by things like substance abuse that hamper systematic planning. People with that level of capability almost always manage to create sufficiently comfortable life situations for themselves that they they lack motivation to do so.

Problems happen when either (1) technology to do harm becomes commoditized and freely available so less capable people can acquire and use (eg. Easy conversion of semi automatic to automatics, commoditized hacks for script kiddies) or (2) capable people feel they're been sufficiently wronged that they don't care about the consequences of revenge (pretty much all the really big headline terrorist actions)

Most intelligence activity on prevention focuses on (2) since pulling off such actions generally requires the capable planners to coordinate with less capable "feet on the ground" who can give things away. However capable people acting alone which happens occasionally is pretty much impossible to stop (Timothy McVeigh, Ted Kazynski, Kaloyev)

VHF radios are relatively easy to locate by direction finding techniques, using for example a doppler direction finding array. The FCC's enforcement division is well-known to possess SUVs with this kind of equipment discreetly installed, although due to their famously limited budget you aren't too likely to see one in the wild. On the other hand, though, many large airports have such equipment permanently installed (e.g. RFEye).

https://www.telegraph.co.uk/travel/destinations/oceania/aust...

It happens rarely, but pilots are also trained to not blindly follow clearances. They are still pilot in command.

The problem with _requiring_ authentication is that it can create a safety issue when communications break down. As a pilot it's not a bad idea to carry a spare handheld radio, or even use a cel phone, when you lose communications.

I was flying once with a buddy of mine who was Pilot-In-Command and we lost (or thought we had lost) our radio on an IFR flight back into the Bay Area. What had happened was the Push-to-Talk button on the yoke had become stuck and we were transmitting the entire time while attempting to diagnose what was wrong. This happened for about five minutes. Thankfully he had a handheld radio in his flight bag, and when we turned it on we figured out really quickly what was going on.

In that scenario, if we didn't have the handheld radio, it would have caused a worse problem.

Pilots do train for communication failures. ATC at most airports has a number you can call to be patched through if your radio stops working and I've heard multiple cases of people having to use it.

In the worst case, there are non-verbal methods of communication that pilots or ATC can employ (ATC can send light signals, pilots rock their wings to signal communication failures, etc.)

Now I'm trying to remember if the radios have TX lights that would help show that you are transmitting...

G1000/G3000 radios do. TX and RX

G1000 will disable tx if you are transmitting for 35 seconds straight I think as a way to mitigate a stuck mic.

Yes, even many old school radios will have something that indicates transmitting or not.

Here's an example of a AA B757 in essentially that exact situation after taking off at JFK: https://www.youtube.com/watch?v=Wcq6KipBOro

Seven six, on the fritz! (Squawk 7600.) There’s an established, cut—and-dry procedure for when an IFR aircraft goes NORDO. Even without that handheld, you would’ve been fine.

Every airplane between you and your destination, on the other hand, would find themselves being rerouted.

I didn't mean to imply that we thought we were in any kind of imminent danger. There were three pilots (including myself) in the plane, all whom had trained in a highly congested area.

My point was that having the handheld radio allowed us to trouble shoot the situation. Without the radio there would have been a lot higher workload and less of a margin for safety.

Plus you might've inadvertently blocked the frequency longer (for other participants) if you hadn't been able to diagnose it. So it was definitely a good thing you had the handheld aboard: even if there was no imminent danger for you, it would have been an inconvenience for many.

With all the instrumentation in a cockpit I'm surprised there isn't a simple light indicator that your radio is currently broadcasting.

Dumb question: how would a cellphone work given they don't work on passenger aircraft?

General aviation flies a lot lower and slower than passenger jets.

I've found that unless I'm under about 10000' and under about 150 knots, cell phones don't work very well in GA either.

Nice plane you have, if you fly above 10000' or above 150 knots frequently :-)

I've upgraded since, but my old 182 would do 10K any day and 150knots with a slight tailwind.

did it have the retractable gear? I've always loved that design.

No, mine was a straight 182. I agree that the 182RG is a great airplane.

I've accidentally had phones on during takeoff and they lost signal within seconds. Were they already going too fast? Or was there jamming?

Doppler shift for the signal?

Yes. Speed

I've managed to keep a data connection pretty far into the initial ascent. It was surprising!

Moving too fast for the cell network to keep up, I'm guessing. Before it manages to handoff to the next cell, you're actually leaving that one already.

> Incidentally, I am happily surprised that that never seems to happen. One could so easily create so much chaos at an airport... There is no authentication whatsoever.

https://en.wikipedia.org/wiki/Airband

One of the reasons why airband radio uses amplitude modulation is so that a much more powerful transmitter (ground based, operated by air traffic control or tower personnel, with access to very high Tx powers), can transmit on top of spurious transmissions coming from low powered aircraft or handheld transmitters. The big ground based Tx can step on an interloper and communicate its message to all aircraft on the same frequency.

> Air traffic controllers have primary radar and secondary radar in addition to ADS-B information.

No they don't.

Most secondary (surveillance) radar in the US has been shut down already.

The FAA has been busy shutting down basically anything that requires maintenance, and LORAN is gone too.

I believe the whole point of ADS-B out was to shut down basically the rest.

Source: commercially-rated pilot.

If you shut it all down and something catastrophic somehow goes wrong with GPS, it will be very difficult to get the thousands of planes in the air back onto runways.

Are they also getting rid of VOR for navigation?

https://en.wikipedia.org/wiki/VHF_omnidirectional_range

(Sorry, paywall is blocking me)

Not too familiar with aviation or IFR. Isn't VOR primarily a beacon that you use to orient yourself, doesn't seem like it helps the ATC maintain separation etc..

IFR separation does not have to depend on radar. Though radar procedures are the norm, there are non-radar procedures wherein ATC can issue instructions (clearances) that provide separation for airplanes who are all navigating themselves with respect to ground-based or satellite fixes. In that regard, VORs help ATC maintain separation.

VOR is used for many things, including navigation, holding and locator beacons.

I suspect VORs will be kept for airport locations and en route instructions, but most others will be shut down in favor of GPS.

VOR navigation is very much still a thing.

They've been reducing the number of redundant ground stations lately as part of a project called "Minimum Operational Network", or "MON". (Mostly on affecting the east coast.) But they're explicitly making sure there's enough coverage that you can get a VOR signal anywhere above 5,000 feet AGL.

More info: https://www.faa.gov/about/office_org/headquarters_offices/at...

Do you have any citations from the FAA for that?

Given that then ADS-B mandate was just a couple weeks ago, it seems logistically odd that secondary radar would already be turned off.

More generally, my understanding was that the FAA was keeping primary+secondary radar around to augment ADS-B via sensor fusion. I know the "reply" light on my plane's radar transponder still lights up regularly, so at least out here on the west coast the secondary radar is still active. :)

I've also heard there's a longer-term plan to consolidate ATC radar into the NOAA weather radar stations, once they upgrade weather radar to phased-arrays at some unknown point in the future. The marginal cost to keep secondary radar running in that environment is basically zero.

That said, they have certainly been turning down many of the redundant VOR stations as part of MON, and they just turned HIWAS off last week.

I don't think anyone really misses LORAN, though. ;)

So then if everything goes to ADS-B and GPS, doesn't that create a potential single-point-of-failure scenario, as isn't the data that ADS-B sends out itself derived from GPS? Having all navigation information come from a single source strikes me as about a good idea as the single AoA sensor on the 737MAX.

I'd like to be wrong about this, but I'm not sure how I am. How am I potentially wrong about this?

You're correct. That's the direction the FAA is headed.

It doesn't have to make sense.

Secondary radar is very much alive in the US. Forget to turn on your transponder (whether or not you are ADS-B out equipped) and how long until ATC notices? For me, it's usually been before I'm through 400' AGL. Or, as tjohns mentions, just look at how often your reply light flashes.

From the FAA itself: "The Mode S SLEP Phase 2 program will implement modifications to the Mode S system to sustain secondary surveillance service through 2028."

0 - https://www.faa.gov/air_traffic/publications/cip/files/FY16-...

Err... The FAA is replacing all their secondary Mode S radars with... secondary Mode S radars with ADS-B capability. That's probably gonna last at least 20 years, once installed.

Maybe you're thinking of old-school SSR radars ?

In addition to the systems you mentioned, many commercial aircraft are equipped with an Inertial Navigation System[0].

[0] https://aviation.stackexchange.com/questions/43283/do-todays...

> Air traffic controllers have primary radar and secondary radar in addition to ADS-B information.

Article is pay-walled so I didn't read, but the title explicitly says 'from Radar to GPS', ie away from Radar altogether. It doesn't say, GPS to supplement radar.

Reply on news.ycombinator.com