6 ms·

Yikes. Was kinda hoping to see it just be an NFC sticker you have to slap on your drone after registration, but this reads like they want an active radio and pe

Yikes. Was kinda hoping to see it just be an NFC sticker you have to slap on your drone after registration, but this reads like they want an active radio and per-flight generated IDs, the devices need internet connections and all kinds of business that precludes entire categories of lightweight UASes. This is definitely going to be stifling to the market if it's enforced in the slightest. It also rules out using ADS-B on drone aircraft entirely, which is a bit perplexing.

sixdimensional 7y ago
Did you hear that they are having privacy concerns for the 24 bit ICAO aircraft address which is broadcast by ADS-B transponders now?? There are a bunch of companies petitioning for a temporary anonymous private address process that only the FAA would be able to track. Why did they bother to invent a publicly broadcast address if it wasn't intended to be used that way!?
- Maxious 7y ago
  > The Privacy ICAO Address (PIA) program is now available. https://www.faa.gov/nextgen/equipadsb/privacy/ https://www.faa.gov/nextgen/equipadsb/privacy/
  - blantonl 7y ago
    This will be easily defeatable with a crowd-sourced database. It is not even close to easy to be able to change an ADS-B out ICAO address in avionics right now, so periodically changing these addresses won't happen in production. When will people realize that security by obscurity doesn't work?
- tjohns 7y ago
  ADS-B has always had an "anonymous" option for aircraft using the UAT uplink, as long as they're squawking 1200/VFR. (Similarly, there's long been a way to request that specific aircraft are hidden from the FAA's public radar feeds.) Aircraft using the 1090ES uplink can't take advantage of the privacy mode, because it's based on top of Mode S, which is an older protocol and never had privacy built in - it didn't need it before because Mode S just didn't include position data. (We'll ignore Mode S multilateration, which is a recent development.) The private ICAO addresses are a hack to provide similar functionality as UAT. > Why did they bother to invent a publicly broadcast address if it wasn't intended to be used that way!? Because those addresses were invented for Mode S, which didn't broadcast position. It was part of the secondary radar (transponder) system, and needed to be paired with primary radar to get something useful.
ars 7y ago
Sounds to me like you need to submit a comment. Seriously, your viewpoint and knowledge is EXACTLY what they want to hear.
ohazi 7y ago
Seems weird to require both an internet connection and radio broadcasting from the UAV in operation, but that seems to be what the rule states. Why not either-or?
- anticensor 7y ago
  The rule is specifically designed to track people attempting to fly civilian UAVs in military zones.
- jpalomaki 7y ago
  Internet connection would allow some automation to educate users of no-fly zones. For example send notification emails afterwards. Local broadcast might make it easier to identify drones which are behaving badly. For example if you have many drones operating in area, maybe you could catch the ID of specific drone using directional antenna.
  - ethbro 7y ago
    There's finite available RF bandwidth, specifically in long-life things like air systems where you can't just add additional bands decades down the road. Ergo, trying to keep the most numerous systems (smaller UAVs) from needing to actively broadcast seems like a good idea. Remember that one of the reasons cell phones were originally prohibited from planes was due to signal propagation (back when they used higher power to broadcast). Frequency isn't infinite, and a 250' tall transmitter covers a very different amount of area than one at ground level.
pininja 7y ago
ADS-B, while good, isn’t perfect. It’s refresh rate goes down the more aircraft there are in an area to avoid signal conflict, it’s one way broadcast so there isn’t a way to request a different view of data, and it’s more like an untrusted client (self reporting) with no civilian ability to support cryptographic identity (no way to prove the signal is real, authentic, or correct). Remote ID probably doesn’t solve all of these issues, but it may try to solve some around identification. I’m not familiar enough with the specification yet or how it’ll be implemented. I work on Uber Elevate, which has been an industry partner for testing Remote ID. https://medium.com/wing-aviation/wing-partners-with-industry-and-regulators-to-validate-the-remote-identification-of-drones-based-252e98e1c7a4 https://medium.com/wing-aviation/wing-partners-with-industry...
- ihuman 7y ago
  What do you mean by "a different view of data"?
  - pininja 7y ago
    With ADS-B the broadcast packet contains whatever it contains (a consumer cannot request different “endpoints”). By different view I was thinking of how an internet based solution could have endpoints with many options and pull too instead of only push.
- mike_d 7y ago
  > it’s one way broadcast so there isn’t a way to request a different view of data I think what you mean is there isn't a way for _you_ to request data. I'm sure you are operating a SDR on 1090 MHz as an ADS-B receiver, but that is only half the equation. Aircraft transponders also listen on 1030 MHz for interrogations and will respond appropriately. Secondary surveillance radar does exactly this in the United States around towered airports. ADS-B is just a dumbed down version of the Identify Friend or Foe system deployed by the military. > I work on Uber Elevate, which has been an industry partner for testing Remote ID Fantastic! If you'd like to grab coffee and learn more about how this stuff works (and how it should work in the future), my email is in my profile.
CrazyStat 7y ago
The internet connection doesn't have to be on the UAS itself, it can be through a paired smartphone or something. Also if the UAS has a radio then it only needs to connect to the internet once at the beginning of the flight. There are concerns about overwhelming the ADS-B spectrum with a bunch of drones. It's not ruled out entirely--there are exceptions--but they want to avoid a whole bunch of drones using it, yes.
- ethbro 7y ago
  All the outrage on here seems shortsighted, while the FAA and you are recognizing the long game. There's no putting the genie back in the bottle, so any proposal needs to scale across the widest range of UAS, while coping with potentially exponential vehicle count growth, while avoiding adverse impacts to any legacy systems (existing ATC instruments or RF networks). That's no trivial circle to square. By offloading performance demands onto clearing houses, they gain scalability. By avoiding existing devices, they avoid negatively impacting legacy traffic. I'd imagine existing flight instruments wouldn't be too happy if I 100x'd the amount of data they're receiving (assuming the spectrum would even support it). So yes, additional requirements, but that's how you get to a world where there are safely 1000s of drones in any given airspace.
mdip 7y ago
My dad was a licensed pilot (IFR with several other certifications, but small plane). My understanding of how the FAA handles regulations is that they tend to be onerous. I skimmed the doc, but going by that understanding, I'm not at all surprised that they'd propose a regulation that would effectively eliminate entire products because they'd be impossible to build in a compliant manner. The law of unintended consequences lands pretty hard on government regulations -- mini-motorcycles come to mind; they don't fall into a category that can be made legal to use anywhere other than private property. And while there are certainly many products where "that's just fine", when things fall into that category, it makes it impossible/extremely expensive to get any form of insurance which could also be required in order to operate the thing in the first place[0]. If what you've written is true, without caveats, it'll be stifling to the market if it is has any chance of being enforced at all. A good chunk of the drone market is hobbyist and a lot of hobbyists have more hobbies than they have time. Unless they also have a regulatory compliance hobby to go along with it, a lot of them are likely to just give up on the whole thing. Of course, this benefits makers of higher-priced drones meant for commercial markets. Some of the inexpensive drones are capable enough for professionals (or can be modded to be made capable enough) -- that's eating into sales of the $1000 -- probably far better -- drones. I'm not saying DJI or some of the larger drone makers are trying to game the system to eliminate smaller competitors, but it wouldn't really surprise me. The things they're expecting drone manufacturers to implement would require lower-priced drones to raise their prices to offset the cost. The more expensive drones might be able to cheapen-up elsewhere to offset, or they'll similarly pass that on to the consumer since bumping the price of a $1000 drone to $1050 would barely be noticed vs. a $150 drone costing $200 after being made compliant[1]. [0] Minibikes are a bad example, but the idea is that people will be less willing to buy/use a product (that requires skill to operate safely) if they cannot insure against a lawsuit when it hurts someone/causes some other form of harm that a lawyer can profitably latch onto. [1] Hypothetical prices, obviously -- I don't understand everything that will be required nor who would be required to comply. I.e. If you can't sell non-compliant models, it'll bump up the price. If you can, it'll bump up the cost of ownership for those who aren't interested in explaining themselves to a judge/officer.
- bsder 7y ago
  > I skimmed the doc, but going by that understanding, I'm not at all surprised that they'd propose a regulation that would effectively eliminate entire products because they'd be impossible to build in a compliant manner. Well, given that we continue to have fire control flights grounded because of idiots flying drones, the FAA is going to keep ratcheting up the regulations until that stops. So, either people can start getting on board with some tracking and actually provide some useful feedback and advice, or the FAA is eventually just going to ban all drones and make it a major felony to possess one.

sixdimensional 7y ago

Did you hear that they are having privacy concerns for the 24 bit ICAO aircraft address which is broadcast by ADS-B transponders now?? There are a bunch of companies petitioning for a temporary anonymous private address process that only the FAA would be able to track. Why did they bother to invent a publicly broadcast address if it wasn't intended to be used that way!?

Maxious 7y ago
> The Privacy ICAO Address (PIA) program is now available. https://www.faa.gov/nextgen/equipadsb/privacy/ https://www.faa.gov/nextgen/equipadsb/privacy/
- blantonl 7y ago
  This will be easily defeatable with a crowd-sourced database. It is not even close to easy to be able to change an ADS-B out ICAO address in avionics right now, so periodically changing these addresses won't happen in production. When will people realize that security by obscurity doesn't work?
tjohns 7y ago
ADS-B has always had an "anonymous" option for aircraft using the UAT uplink, as long as they're squawking 1200/VFR. (Similarly, there's long been a way to request that specific aircraft are hidden from the FAA's public radar feeds.) Aircraft using the 1090ES uplink can't take advantage of the privacy mode, because it's based on top of Mode S, which is an older protocol and never had privacy built in - it didn't need it before because Mode S just didn't include position data. (We'll ignore Mode S multilateration, which is a recent development.) The private ICAO addresses are a hack to provide similar functionality as UAT. > Why did they bother to invent a publicly broadcast address if it wasn't intended to be used that way!? Because those addresses were invented for Mode S, which didn't broadcast position. It was part of the secondary radar (transponder) system, and needed to be paired with primary radar to get something useful.

ars 7y ago

Sounds to me like you need to submit a comment. Seriously, your viewpoint and knowledge is EXACTLY what they want to hear.

ohazi 7y ago

Seems weird to require both an internet connection and radio broadcasting from the UAV in operation, but that seems to be what the rule states. Why not either-or?

anticensor 7y ago
The rule is specifically designed to track people attempting to fly civilian UAVs in military zones.
jpalomaki 7y ago
Internet connection would allow some automation to educate users of no-fly zones. For example send notification emails afterwards. Local broadcast might make it easier to identify drones which are behaving badly. For example if you have many drones operating in area, maybe you could catch the ID of specific drone using directional antenna.
- ethbro 7y ago
  There's finite available RF bandwidth, specifically in long-life things like air systems where you can't just add additional bands decades down the road. Ergo, trying to keep the most numerous systems (smaller UAVs) from needing to actively broadcast seems like a good idea. Remember that one of the reasons cell phones were originally prohibited from planes was due to signal propagation (back when they used higher power to broadcast). Frequency isn't infinite, and a 250' tall transmitter covers a very different amount of area than one at ground level.

pininja 7y ago

ADS-B, while good, isn’t perfect. It’s refresh rate goes down the more aircraft there are in an area to avoid signal conflict, it’s one way broadcast so there isn’t a way to request a different view of data, and it’s more like an untrusted client (self reporting) with no civilian ability to support cryptographic identity (no way to prove the signal is real, authentic, or correct). Remote ID probably doesn’t solve all of these issues, but it may try to solve some around identification. I’m not familiar enough with the specification yet or how it’ll be implemented. I work on Uber Elevate, which has been an industry partner for testing Remote ID. https://medium.com/wing-aviation/wing-partners-with-industry-and-regulators-to-validate-the-remote-identification-of-drones-based-252e98e1c7a4 https://medium.com/wing-aviation/wing-partners-with-industry...

ihuman 7y ago
What do you mean by "a different view of data"?
- pininja 7y ago
  With ADS-B the broadcast packet contains whatever it contains (a consumer cannot request different “endpoints”). By different view I was thinking of how an internet based solution could have endpoints with many options and pull too instead of only push.
mike_d 7y ago
> it’s one way broadcast so there isn’t a way to request a different view of data I think what you mean is there isn't a way for _you_ to request data. I'm sure you are operating a SDR on 1090 MHz as an ADS-B receiver, but that is only half the equation. Aircraft transponders also listen on 1030 MHz for interrogations and will respond appropriately. Secondary surveillance radar does exactly this in the United States around towered airports. ADS-B is just a dumbed down version of the Identify Friend or Foe system deployed by the military. > I work on Uber Elevate, which has been an industry partner for testing Remote ID Fantastic! If you'd like to grab coffee and learn more about how this stuff works (and how it should work in the future), my email is in my profile.

CrazyStat 7y ago

The internet connection doesn't have to be on the UAS itself, it can be through a paired smartphone or something. Also if the UAS has a radio then it only needs to connect to the internet once at the beginning of the flight. There are concerns about overwhelming the ADS-B spectrum with a bunch of drones. It's not ruled out entirely--there are exceptions--but they want to avoid a whole bunch of drones using it, yes.

ethbro 7y ago
All the outrage on here seems shortsighted, while the FAA and you are recognizing the long game. There's no putting the genie back in the bottle, so any proposal needs to scale across the widest range of UAS, while coping with potentially exponential vehicle count growth, while avoiding adverse impacts to any legacy systems (existing ATC instruments or RF networks). That's no trivial circle to square. By offloading performance demands onto clearing houses, they gain scalability. By avoiding existing devices, they avoid negatively impacting legacy traffic. I'd imagine existing flight instruments wouldn't be too happy if I 100x'd the amount of data they're receiving (assuming the spectrum would even support it). So yes, additional requirements, but that's how you get to a world where there are safely 1000s of drones in any given airspace.

mdip 7y ago

My dad was a licensed pilot (IFR with several other certifications, but small plane). My understanding of how the FAA handles regulations is that they tend to be onerous. I skimmed the doc, but going by that understanding, I'm not at all surprised that they'd propose a regulation that would effectively eliminate entire products because they'd be impossible to build in a compliant manner. The law of unintended consequences lands pretty hard on government regulations -- mini-motorcycles come to mind; they don't fall into a category that can be made legal to use anywhere other than private property. And while there are certainly many products where "that's just fine", when things fall into that category, it makes it impossible/extremely expensive to get any form of insurance which could also be required in order to operate the thing in the first place[0]. If what you've written is true, without caveats, it'll be stifling to the market if it is has any chance of being enforced at all. A good chunk of the drone market is hobbyist and a lot of hobbyists have more hobbies than they have time. Unless they also have a regulatory compliance hobby to go along with it, a lot of them are likely to just give up on the whole thing. Of course, this benefits makers of higher-priced drones meant for commercial markets. Some of the inexpensive drones are capable enough for professionals (or can be modded to be made capable enough) -- that's eating into sales of the $1000 -- probably far better -- drones. I'm not saying DJI or some of the larger drone makers are trying to game the system to eliminate smaller competitors, but it wouldn't really surprise me. The things they're expecting drone manufacturers to implement would require lower-priced drones to raise their prices to offset the cost. The more expensive drones might be able to cheapen-up elsewhere to offset, or they'll similarly pass that on to the consumer since bumping the price of a $1000 drone to $1050 would barely be noticed vs. a $150 drone costing $200 after being made compliant[1]. [0] Minibikes are a bad example, but the idea is that people will be less willing to buy/use a product (that requires skill to operate safely) if they cannot insure against a lawsuit when it hurts someone/causes some other form of harm that a lawyer can profitably latch onto. [1] Hypothetical prices, obviously -- I don't understand everything that will be required nor who would be required to comply. I.e. If you can't sell non-compliant models, it'll bump up the price. If you can, it'll bump up the cost of ownership for those who aren't interested in explaining themselves to a judge/officer.

bsder 7y ago
> I skimmed the doc, but going by that understanding, I'm not at all surprised that they'd propose a regulation that would effectively eliminate entire products because they'd be impossible to build in a compliant manner. Well, given that we continue to have fire control flights grounded because of idiots flying drones, the FAA is going to keep ratcheting up the regulations until that stops. So, either people can start getting on board with some tracking and actually provide some useful feedback and advice, or the FAA is eventually just going to ban all drones and make it a major felony to possess one.