Unencrypted ADS-B Out

[denverpilot]

I've been saying that it should be encrypted and authenticated via PKI... for years now.

From Twitter...

“@CS_Intel: AN FBI CESSNA 206, REG N251SR, IS CURRENTLY TRAVERSING THE BOSTON-METRO AREA AT 3,400 FT ASL. ADS-B CNFRM'D.”

Now hey, we all know they could hide their N-number and yadda yadda, but this shows the direct reception of the ADS-B Out signal is a privacy problem...

How long until someone sets up an Internet database and accepts feeds from hobbyists of all ADS-B transmissions in real-time?

Wheee! Bad data comm designs are fun! LOL...

 

[N801BH]

I've been saying that it should be encrypted and authenticated via PKI... for years now.

From Twitter...

“@CS_Intel: AN FBI CESSNA 206, REG N251SR, IS CURRENTLY TRAVERSING THE BOSTON-METRO AREA AT 3,400 FT ASL. ADS-B CNFRM'D.”

Now hey, we all know they could hide their N-number and yadda yadda, but this shows the direct reception of the ADS-B Out signal is a privacy problem...

How long until someone sets up an Internet database and accepts feeds from hobbyists of all ADS-B transmissions in real-time?

Wheee! Bad data comm designs are fun! LOL...

Wait till you get home and find a Ford LTD with tinted windows parked down the road...

 

[MAKG1]

You mean like flightaware does for IFR traffic?

Since when is an aircraft position private or secret?

Frankly, I want other pilots in nearby airspace to know where I am, and I don't care much about others.

What I do care about is the possibility of injecting false data into the system.

If you encrypted it, what would keep a surveillance system from taking an ADS-B receiver outside an airplane and broadcasting that on the Internet? Seems like you attacked the wrong problem here.

 

[murphey]

Wait till you get home and find a Ford LTD with tinted windows parked down the road...

nah, they're using Suburbans...Ford discontinued the LTD some time ago.

 

[John Collins]

I've been saying that it should be encrypted and authenticated via PKI... for years now.

From Twitter...

“@CS_Intel: AN FBI CESSNA 206, REG N251SR, IS CURRENTLY TRAVERSING THE BOSTON-METRO AREA AT 3,400 FT ASL. ADS-B CNFRM'D.”

Now hey, we all know they could hide their N-number and yadda yadda, but this shows the direct reception of the ADS-B Out signal is a privacy problem...

How long until someone sets up an Internet database and accepts feeds from hobbyists of all ADS-B transmissions in real-time?

Wheee! Bad data comm designs are fun! LOL...

You mean like this? http://www.scanningtasmania.org/adsbfeed.php

 

[gismo]

You mean like flightaware does for IFR traffic?

Since when is an aircraft position private or secret?

Frankly, I want other pilots in nearby airspace to know where I am, and I don't care much about others.

But there are plenty with legitimate reasons for hiding from that and there's a system in place that provides two ways to hide from flightaware. Nothing is going to allow you to hide with ADS-B except turning off and what I don't like is any incentive to do that.

What I do care about is the possibility of injecting false data into the system.

If you encrypted it, what would keep a surveillance system from taking an ADS-B receiver outside an airplane and broadcasting that on the Internet? Seems like you attacked the wrong problem here.

Agreed. Since all planes need to have the ability to decode the transmissions, encrypting them to prevent interception of the info isn't likely to be feasible. But similar problems exist with any scheme that precludes spoofing. Fortunately true spoofing (fooling an airplane into thinking it's somewhere it's not) is far more complex and difficult than simply jamming and for the most part can only target one airplane at a time effectively.

 

[denverpilot]

Each aircraft gets their own traffic warnings encrypted to them specifically, if it had been designed with enough bandwidth to do so. No need for any transmission to be public.

I haven't looked. Are military aircraft required to participate in ADS-B? Will everyone be receiving their locations in the clear from ground station uplinks if they're within the ADS-B hockey puck?

(Obviously I'm sure they've written procedures to stop participating at any particular time, but will they participate in high traffic areas during peacetime ops?)

 

[Bob Noel]

But similar problems exist with any scheme that precludes spoofing. Fortunately true spoofing (fooling an airplane into thinking it's somewhere it's not) is far more complex and difficult than simply jamming and for the most part can only target one airplane at a time effectively.

"far more complex"? I don't think so. Creating false reports is trivial. And since there is absolutely no validation of ADS-B messages, there is no way to detect false reports.

For those that think correlating radar targets with ADS-B message could be used: sure, but if there is radar coverage, why bother with the expense of ADS-B? think it through.

 

[n12365]

I've been saying that it should be encrypted and authenticated via PKI... for years now.

From Twitter...

“@CS_Intel: AN FBI CESSNA 206, REG N251SR, IS CURRENTLY TRAVERSING THE BOSTON-METRO AREA AT 3,400 FT ASL. ADS-B CNFRM'D.”

Now hey, we all know they could hide their N-number and yadda yadda, but this shows the direct reception of the ADS-B Out signal is a privacy problem...

How long until someone sets up an Internet database and accepts feeds from hobbyists of all ADS-B transmissions in real-time?

Wheee! Bad data comm designs are fun! LOL...

I wonder if the 206 was using a 1090ES transponder or a UAT to transmit ADS-B messages. The 1090ES messages are just expanded Mode-S messages containing ADS-B position date. The UAT setup is a separate solution and allows for anonymous ADS-B messages. Other aircraft are able to see your position, but they don't necessarily know who you are. I would guess they were using 1090ES. There are a wide variety of tools available for receiving and decoding 1090ES, but not for UATs which are US specific.

All it takes is a PC, a $20 software defined radio, and some simple software to capture ADS-B messages from the ground. I am currently doing that as part of my graduate capstone project at Embry-Riddle. Once you have an aircraft's mode-S address, all sorts of data is available via the FAA's aircraft registry. Btw, about 26% of aircraft in the Washington DC area using Mode-S transponders are transmitting ADS-B position data.

Ryan

 

[John Collins]

For those that think correlating radar targets with ADS-B message could be used: sure, but if there is radar coverage, why bother with the expense of ADS-B? think it through.

The plan is to keep most of the radars as a backup system for ADS-B, that is why transponders are still required. ADS-B won't save much money as a result of reducing the number of radars. ADS-B offers the potential of closer spacing between aircraft and other services that radar can't offer.

 

[N801BH]

The plan is to keep most of the radars as a backup system for ADS-B, that is why transponders are still required. ADS-B won't save much money as a result of reducing the number of radars. ADS-B offers the potential of closer spacing between aircraft and other services that radar can't offer.

ADS-B if for positive identification of each aircraft so the guv can send each registered owner a bill for using the airspace....... If you think otherwise, then I have some Kool-Aid for non believers to drink..

 

[John Collins]

ADS-B if for positive identification of each aircraft so the guv can send each registered owner a bill for using the airspace....... If you think otherwise, then I have some Kool-Aid for non believers to drink..

Well, then installing a UAT and always flying anonymous will keep the cool aid from your lips.

 

[jtheune]

I do recall from a old Heinlein story where the hero , when taking off switched his transponder to a code that belonged to a another person so he could hide his identity. Wonder when that piece of SF will become true?

 

[Anymouse]

I'm not sure about the current configuration, but when I was flying with the "original" ADS-B as part of the initial Capstone Program in Alaska, we had the option to turn off all identifying information. All that would be broadcast was a simple "VFR" target.

 

[Bob Noel]

The plan is to keep most of the radars as a backup system for ADS-B, that is why transponders are still required. ADS-B won't save much money as a result of reducing the number of radars. ADS-B offers the potential of closer spacing between aircraft and other services that radar can't offer.

Believe me, I know what could be done with ADS-B. It could be much better at collosion avoidance than TCA, providing horizontal conflict resolution instead of just vertical.

The biggest problem with ADS-B is that lack of message validation.

ok, another problem is the gold-plating of the specifications (classic requirements by committee).

 

[gismo]

"far more complex"? I don't think so. Creating false reports is trivial. And since there is absolutely no validation of ADS-B messages, there is no way to detect false reports.

We're talking about two different things. As you say, generating false targets is easy, eliminating legitimate ones not so much, and moving legitimate ones even harder. I also agree that there's no way to validate targets aside from correlation with transponder returns and that it's unlikely to be implemented.

 

[gismo]

Each aircraft gets their own traffic warnings encrypted to them specifically, if it had been designed with enough bandwidth to do so. No need for any transmission to be public.

That would eliminate one of the useful functions of ADS-B, the ability to "see" other nearby aircraft without support from the ground. In addition it would require all aircraft receiving traffic info from ground stations to be uniquely identified so being anonymous would be difficult.

I haven't looked. Are military aircraft required to participate in ADS-B? Will everyone be receiving their locations in the clear from ground station uplinks if they're within the ADS-B hockey puck?

(Obviously I'm sure they've written procedures to stop participating at any particular time, but will they participate in high traffic areas during peacetime ops?)

I've heard opposing views on military participation but I think it's very likely they will be required to equip for ADS-B out by 2020 if they operate in US airspace like the rest of us. Certainly there will be a "steath" option for use when needed.

 

[denverpilot]

That would eliminate one of the useful functions of ADS-B, the ability to "see" other nearby aircraft without support from the ground. In addition it would require all aircraft receiving traffic info from ground stations to be uniquely identified so being anonymous would be difficult.

Once everyone's participating in ADS-B Out (which was the *design* all along, remember... which is what I'm poking at... The design being flawed...) that's exactly what you have.

The incidental ability to "see" stuff around you wasn't specifically listed as a design requirement that I've seen documented.

Portions of TCAS accomplish the same end goal without more than a transponder serial number which doesn't even have to be accurately programmed to still have the system function. Will ADS-B filter out aircraft transmissions from N-numbers not found in the database, or are there rules against transmitting an inaccurate N-number?

We could all be transmitting a random fake N-number or an encrypted version of our N-number and the system still works.

That'd keep the onus on controllers to change the data block for their convenience, and requires no actual identifying information be encoded in the clear over the air.

Since that's still a requirement for positive control anyway, what's the point of ADS-B Out transmitting a real N-number unencrypted? Only FAA needs that information.

Another aircraft avoiding me or vice-versa doesn't need to know who I am, just where I am.

 

[MAKG1]

ADS-B if for positive identification of each aircraft so the guv can send each registered owner a bill for using the airspace....... If you think otherwise, then I have some Kool-Aid for non believers to drink..

You don't need ADS-B for that. It could be done with existing transponders. Time to lose the tinfoil.

 

[gismo]

You don't need ADS-B for that. It could be done with existing transponders. Time to lose the tinfoil.

Possible with Mode S, not with Mode C transponders. And unlike ADS-B out Mode S isn't required. I'm thinking my eventual ADS-B implementation will need UAT as well as Mode S-ES with the ability to disable the Mode S. Or I might just give up on flight above FL 180 since I don't really need that ability.

 

[Palmpilot]

What's "UAT"?

 

[Bob Noel]

What's "UAT"?

Universal Access Transceiver.

Developed for the low-end market that didn't need the added expenses of Mode-S.

 

[Palmpilot]

How does that enable flying anonymously, as John suggested?

 

[Bob Noel]

How does that enable flying anonymously, as John suggested?

Don't know. You'd have to look at the mesage set for UAT. I don't have the standard handy.

Sorry.

 

[Palmpilot]

Maybe John will see my question and explain. Thanks for the info.

 

[n12365]

How does that enable flying anonymously, as John suggested?

When flying VFR, UATs have a feature that allow you to anonymously transmit your location information. Still allows for ATC surveillance and for other aircraft to see you as traffic, but no one knows who you are. It is the equivalent of squawking 1200. Take a look at the ADS-B final rule and search for anonymous http://www.gpo.gov/fdsys/pkg/FR-2010-05-28/pdf/2010-12645.pdf

The 1090ES option is an expansion of a standard Mode-S transponder, so your Mode-S id is being transmitted with each message. The Mode-S id is tied to your N number, so this lets everyone identify who you are. Unlike UATs, there is no anonymous feature.

Ryan

 

[n12365]

Don't know. You'd have to look at the mesage set for UAT. I don't have the standard handy.

Unfortunately, the standards aren't freely available. The FARs relating to ADS-B incorporate standards developed by the RTCA, which are copyrighted documents. You can either pay to get each individual standard or join the RTCA, which starts at $600 per year. The FAA isn't the only government agency that does this. It is a terrible policy to have government regulations enforce standards that aren't available for free, but I don't see this changing anytime soon.

Ryan

 

[Piloto]

When flying VFR, UATs have a feature that allow you to anonymously transmit your location information. Still allows for ATC surveillance and for other aircraft to see you as traffic, but no one knows who you are. It is the equivalent of squawking 1200. Take a look at the ADS-B final rule and search for anonymous http://www.gpo.gov/fdsys/pkg/FR-2010-05-28/pdf/2010-12645.pdf

The 1090ES option is an expansion of a standard Mode-S transponder, so your Mode-S id is being transmitted with each message. The Mode-S id is tied to your N number, so this lets everyone identify who you are. Unlike UATs, there is no anonymous feature.

Ryan

But, but Ryan you need to have all these ID capability so your wife can keep track of your wherabouts, specially when you are by yourself or with some company. I am sure she would not object about this ADS-B feature.

José

 

[denverpilot]

When flying VFR, UATs have a feature that allow you to anonymously transmit your location information. Still allows for ATC surveillance and for other aircraft to see you as traffic, but no one knows who you are. It is the equivalent of squawking 1200. Take a look at the ADS-B final rule and search for anonymous http://www.gpo.gov/fdsys/pkg/FR-2010-05-28/pdf/2010-12645.pdf

Sounds like a manufacturer should implement it and tout it as a feature.

Know any manufacturers of current UAT gear who do?

The 1090ES option is an expansion of a standard Mode-S transponder, so your Mode-S id is being transmitted with each message. The Mode-S id is tied to your N number, so this lets everyone identify who you are. Unlike UATs, there is no anonymous feature.

Somewhat incorrect in that technically the N-number to encoding is obfuscated by not being a direct link and done by the avionics shop at installation time.

It's not a direct encoding of the N-number other than by implementation, but is not a hard requirement of the Mode-S spec. Mode-S of course, is soon to be required.

It's the difference between "let's all agree to program stuff to encode a code that FAA can relate back to an N-number" and "this box transmits your N-number in the clear". A minor distinction but an important one in data security design.

 

[Jim_R]

Somewhat incorrect in that technically the N-number to encoding is obfuscated by not being a direct link and done by the avionics shop at installation time.

There's hardly any obfuscation. The avionics shop may make the link in the equipment, but the FAA has already made the assignment to your N-number, even if you don't have Mode-S equipment. The Mode-S ID is freely visible on the FAA's public N-number registry records.

 

[N801BH]

There's hardly any obfuscation. The avionics shop may make the link in the equipment, but the FAA has already made the assignment to your N-number, even if you don't have Mode-S equipment. The Mode-S ID is freely visible on the FAA's public N-number registry records.

 

[n12365]

Somewhat incorrect in that technically the N-number to encoding is obfuscated by not being a direct link and done by the avionics shop at installation time.

It's not a direct encoding of the N-number other than by implementation, but is not a hard requirement of the Mode-S spec. Mode-S of course, is soon to be required.

It's the difference between "let's all agree to program stuff to encode a code that FAA can relate back to an N-number" and "this box transmits your N-number in the clear". A minor distinction but an important one in data security design.

Never claimed the N-number is transmitted. The mode-S address and the N-number are tied together by the aircraft's entry in the FAA Aircraft Registry.

 

[N801BH]

Never claimed the N-number is transmitted. The mode-S address and the N-number are tied together by the aircraft's entry in the FAA Aircraft Registry.

Very true...

Here is mine.................

Aircraft Registration» Aircraft Inquiry»N-Number Inquiry




FAA REGISTRY

N-Number Inquiry Results



N801BH is Assigned Data Updated each Federal Working Day at Midnight



Aircraft Description

Serial Number8-4419
StatusValid
Manufacturer NameHAAS POWER AIR INC
Certificate Issue Date05/19/2003ModelHAAS 801
Expiration Date05/31/2014
Type AircraftFixed Wing Single-EngineType EngineReciprocating
Pending Number ChangeNoneDealerNoDate Change
AuthorizedNoneMode S Code (base 8 / oct)52562665
Mode S Code (base 16 / hex)AAE5B5Type
Fractional OwnerNO
Kit ManufacturerZENITH ACFT COKit
ModelZENITH 801


Registered Owner

NameHAAS POWER AIR INC
Street700 PONDEROSA DR
CityJACKSON
StateWYOMING
CountyTETON
Zip Code83001-9379
CountryUNITED STATES


Airworthiness

Engine ManufacturerAMA/EXPR
ClassificationExperimental
Engine ModelUNKNOWN ENG
CategoryAmateur
Built</SPAN> A/W Date03/23/2004

 

[John Collins]

Sounds like a manufacturer should implement it and tout it as a feature.

Know any manufacturers of current UAT gear who do?

All do, it is part of the specification and the rule. The GDL88 supports anonymous as long as the transponder is not mode S. With mode S it always sends the ICAO ID, so the UAT has to as well or you would show up as two targets and this is not permitted. This is one reason not to change a mode A/C transponder to mode S as you will lose anonymity.

Somewhat incorrect in that technically the N-number to encoding is obfuscated by not being a direct link and done by the avionics shop at installation time.

It's not a direct encoding of the N-number other than by implementation, but is not a hard requirement of the Mode-S spec. Mode-S of course, is soon to be required.

In general, the ICAO code is not related to the aircraft registration number, but in the US it is related by an unpublished algorithm. For example, on the GTX330, you can enter the ICAO ID or the N number, the other is calculated and not needed to be entered.

It's the difference between "let's all agree to program stuff to encode a code that FAA can relate back to an N-number" and "this box transmits your N-number in the clear". A minor distinction but an important one in data security design.

The N number is transmitted in the clear text of the mode S message or for a commercial carrier, it can be the flight number that is filed in the flightplan. The ICAO code is exclusive ored with the message parity and transmitted in the parity field.

 

[slacktide]

How long until someone sets up an Internet database and accepts feeds from hobbyists of all ADS-B transmissions in real-time?.

LOL. They already have EXACTLY that. It's called FlightRadar24. http://www.flightradar24.com/increase-coverage

Flightradar24 started as a hobby project in 2006 when two Swedish aviation geeks decided to build a network of ADS-B receivers in Northern and Central Europe. In 2009 we opened up the network, and made it possible for anyone with an ADS-B receiver to upload data to the network. Many parts of the world were quickly covered, but the quest to provide global ADS-B coverage is still ongoing. Hopefully with your support, we will get there.

 

[comanchepilot]

Air Force One uses ADS-B . . . .

Oops. Did I just say that? And its not encrypted.

 

[NineThreeKilo]

ADS-B if for positive identification of each aircraft so the guv can send each registered owner a bill for using the airspace....... If you think otherwise, then I have some Kool-Aid for non believers to drink..

Indeed!

I thought the same thing as they started up again on the usage fees right around the same time as ADS.

I have my plane blocked at the source for flightaware, I don't like my tailnumber broadcast, where I am going, who I am, plane and simple it's no ones business.

Now domestic mil and govt aircraft, I figure the people should be able to track those aircraft, after all we paid for them.

 

[zaitcev]

There are a wide variety of tools available for receiving and decoding 1090ES, but not for UATs which are US specific.

UAT is not U.S.-specific. The problem with UAT is that $35 SDR does not have the sample rate to capture it. To deal with UAT one needs a real SDR like an Ettus. The cheapest of those is $800.

 

[denverpilot]

There's hardly any obfuscation. The avionics shop may make the link in the equipment, but the FAA has already made the assignment to your N-number, even if you don't have Mode-S equipment. The Mode-S ID is freely visible on the FAA's public N-number registry records.

Good point.

I guess we need to know what specific aircraft it is to separate it from others then. (Sarcasm.)

 

[denverpilot]

UAT is not U.S.-specific. The problem with UAT is that $35 SDR does not have the sample rate to capture it. To deal with UAT one needs a real SDR like an Ettus. The cheapest of those is $800.

There's others. Similar price tag, but judging by the pricing on SDR gear that'll be 1/2 that price in six months.