10 ms·

GNSS Jamming and Spoofing – Galileo's Authentication Algorithm Part 3 (2020)

40 points by oritsnile 6y ago

throw0101a 6y ago
> Military solutions will fall back to inertial, celestial or optical guidance, and people using GPS for navigation will at worst show up where they need to be somewhat later than planned. The US Navy re-started celestial navigation a few years ago: * https://www.npr.org/2016/02/22/467210492/u-s-navy-brings-back-navigation-by-the-stars-for-officers https://www.npr.org/2016/02/22/467210492/u-s-navy-brings-bac... This (1960s?) US government produced (45m) video video gives a pretty good overview: * https://www.youtube.com/watch?v=UV1V9-nnaAs https://www.youtube.com/watch?v=UV1V9-nnaAs For those wanting to invest (substantially) more time, the two videos by "Tippecanoe Boats" are slightly rambling at times, but he does lay things out pretty well by the end of it (second is largely examples): * https://www.youtube.com/watch?v=-ARXW8InStY https://www.youtube.com/watch?v=-ARXW8InStY * https://www.youtube.com/watch?v=yu5R5mrrGB0 https://www.youtube.com/watch?v=yu5R5mrrGB0
- PaulHoule 6y ago
  This kind of weapon has a backup INS https://en.wikipedia.org/wiki/Joint_Direct_Attack_Munition https://en.wikipedia.org/wiki/Joint_Direct_Attack_Munition The INS doesn't need to be terribly high performance because it only needs to work for the time it takes a bomb to fall from the sky.
- ardy42 6y ago
  >> Military solutions will fall back to inertial, celestial or optical guidance, and people using GPS for navigation will at worst show up where they need to be somewhat later than planned. > The US Navy re-started celestial navigation a few years ago: Also, the SR-71, B-1, and B-2 and other aircraft have automated celestial navigation systems to provide corrections to their INSs: https://www.thedrive.com/the-war-zone/17207/sr-71s-r2-d2-could-be-the-key-to-winning-future-fights-in-gps-denied-environments https://www.thedrive.com/the-war-zone/17207/sr-71s-r2-d2-cou... Honestly, for military applications, GPS seems to be mainly useful for bad weather and providing navigation to the smallest units (and I'd think a small unit would be able fall back to a map and compass).
  - mycall 6y ago
    ICBMs also use celestial navigation systems for same reasons.
- matheusmoreira 6y ago
  These military instructional videos are so great. They explain a lot of concepts very clearly. Nowadays videos seem to be more casual, with more focus on practice than understanding the theory.
aaron695 6y ago
Here's a good map/report on Russian in action GNSS spoofing - https://www.c4reports.org/aboveusonlystars https://www.c4reports.org/aboveusonlystars
- henriklied 6y ago
  Our news org did an investigation into this as well, and also demonstrated how a simple HackRF could be used to fool most modern GPS receivers: https://nrkbeta.no/2017/09/18/gps-freaking-out-maybe-youre-too-close-to-putin/ https://nrkbeta.no/2017/09/18/gps-freaking-out-maybe-youre-t...
throw0101a 6y ago
The OSNMA protocol discussed is based on Timed Efficient Stream Loss-Tolerant Authentication (TESLA): > This document introduces Timed Efficient Stream Loss-tolerant Authentication (TESLA). TESLA allows all receivers to check the integrity and authenticate the source of each packet in multicast or broadcast data streams. TESLA requires no trust between receivers, uses low-cost operations per packet at both sender and receiver, can tolerate any level of loss without retransmissions, and requires no per-receiver state at the sender. TESLA can protect receivers against denial of service attacks in certain circumstances. Each receiver must be loosely time-synchronized with the source in order to verify messages, but otherwise receivers do not have to send any messages. TESLA alone cannot support non-repudiation of the data source to third parties. * https://tools.ietf.org/html/rfc4082 https://tools.ietf.org/html/rfc4082 * A. Perrig, R. Canetti, J. Tygar and D. Song, “Efficient Authentication and Signing of Multicast Streams over Lossy Channels,” IEEE Symposium on Security and Privacy, pp. 56-73, May 2000.
- oritsnile 6y ago
  Part 1 also has a brief explanation of TESLA: https://berthub.eu/articles/posts/galileos-authentication-algorithm-part-1/ https://berthub.eu/articles/posts/galileos-authentication-al...
MayeulC 6y ago
Is there any cheap, open source alternative to GPS for navigation? I'm thinking of an electronic sextant (or star tracker) for rough position estimation, with a few km accuracy.
- PaulHoule 6y ago
  I'm somewhat annoyed at the accelerometer/gyroscopes that are available to the public, it is like they are all gimped to prevent you from building a weapons guidance system. For instance, the Wiimote doesn't have the dynamic range to handle the highest accelerations you can generate waving your arms, which makes it hard to use that kind of thing for athletic training.
  - brandmeyer 6y ago
    > like they are all gimped to prevent you from building a weapons guidance system. You can buy RLGN off the shelf for a couple thousand bucks. You can buy gyrocompass-capable MEMS off of digi-key for dozens of dollars (Murata). Its just market forces and cost. You don't need super stable IMUs for game controllers.
  - michaelt 6y ago
    They're not 'gimped' by any sort of conspiracy. It's just extremely hard to make a pure INS navigation system. Even if you could measure your acceleration accurate to 0.00001 m/s/s, because it gets integrated twice to give position even that tiny error is going to give you a position that's out by 100m after an hour. And measuring acceleration precise to 0.00001 m/s/s is a demanding task - like weighing a car so precisely you could count the pages of a book on the back seat. That level of precision simply isn't available in the affordable MEMS market segment.
    - generalizations 6y ago
      > They're not 'gimped' by any sort of conspiracy. It bothers me when people go around labeling random stuff conspiracies like that's a legit dismissal. If they were 'gimped', they wouldn't be the only hardware. GPS is still gimped and will shut off over a certain height and speed specifically to prevent use in missiles. And the precision used to be regulated, not by expense, but to give the military the superior tool.
PaulHoule 6y ago
Some people say that there isn't much open literature on GPS anti-spoofing, but there are many patents filed by the likes of Lockheed-Martin, Boeing, BAE, etc. I find the multiple antenna answers interesting. For instance, one of the easier attack scenarios against an airplane is to have a directional antenna on the ground. Because airplanes broadcast their GPS position via ADS-B, you could also know that you'd succeeded. In a case like that, however, the radio signal from the ground would be stronger than the signal from the sky and it would be obvious what was going on, unless the attacker managed to get the power level just right. With multiple receivers you also will see very different results with spoofing than with a real signal. For instance if you had a receiver at the front of the airplane and one at the back of the airplane, the time delay for all the fake satellites would be the same (they all come from the same place) whereas the time delays (e.g. position) would be noticeably different from real sats.
- londons_explore 6y ago
  if an attacker knows that the plane has two antennas, and their exact locations, he/she can generate the "correct" signals at each plane antenna with two attacker antennas and lots of math. Getting their exact locations is simplified by the fact the location is being transmitted by ADS-B...
  - knorker 6y ago
    Nice. That sounds tricky though. You need to be very precise. The attacker needs to provide different received signal to two antennas meters (at most!) apart, at a distance of maybe kilometers, on a moving target. That's a hair thin angle. The aircraft just needs to TDoA "did it come from above or below (assuming an attack from the ground). That's 180 degrees. Or am I missing something? Unless I am, this doesn't sound feasible to me.
    - PaulHoule 6y ago
      The equipment I have pictured in my mind is one of the two-axis trackers that radio hams use to track satellites with a long but narrow Yagi-Uda antenna. (If that can handle the bandwidth) These tend to move in jerks and will get in real trouble if you try to move them over the zenith, but they do a great job with LEO satellites and would do OK to uplink one signal to an airplane. You might be able to hit two receivers if you had a phased array antenna like the Starlink antenna but bigger, but now it isn't a simple hacking project anymore. BTW, don't try it. There are certain things like aviation and nuclear power that "Posse Comitatus" doesn't apply to and you could find yourself looking down the barrel of an M4 carbine and getting frog-talk from the USMC much quicker than you'd expect.
knorker 6y ago
So if I understand this right the signature is 32bits times 24 times 15 per full frame? After the fact, why exactly can I not precalculate my spoofed data stream? I just need to spend: 3*2^31*24*15*(spoof_seconds/30) ops (on average)? (assuming 3 streams for position data) Sure, not cheap, but hardly hard even for even a hobbyist. So ~2^45 ops to spoof 10 minutes of data? That's doable. Is my math off? I have more faith in the direction finding aspects. Here's from the article, an understatement of the year: > To beat these simple tricks, a spoofer will need to have multiple transmitters that actually show the same parallax as the actual satellites. However, you can only do this by placing your transmitters next to the satellites - in space. This raises the bar significantly.
- MauranKilom 6y ago
  The HN formatting is eating your * signs...
  - knorker 6y ago
    Thanks. Fixed.
knorker 6y ago
Another question for knowledgeable people here: Can I not just run two GNSS receivers 10 meters apart (on a ship), and if they report as having the same position, then I know someone is spoofing? It should be really hard to beamform the spoofing to spoof two different locations, at least from a distance because of the precise angles needed. Yes, this assumes that my real GNSS signal is good enough that they normally are 10M apart.
- PaulHoule 6y ago
  There's a patent on that one.
- jaywalk 6y ago
  The article addresses that scenario, and confirms that using multiple receivers would allow you to reliably detect spoofing. They'd report the same location but a slightly different time.
  - knorker 6y ago
    Yes, but such countermeasures (like the ones mentioned in the article) are about one receiver doing fancy RF tricks. What I'm proposing is something entirely done by two cheap and existing receivers today, by the end user.

40 points by oritsnile 6y ago

> Military solutions will fall back to inertial, celestial or optical guidance, and people using GPS for navigation will at worst show up where they need to be somewhat later than planned. The US Navy re-started celestial navigation a few years ago: * https://www.npr.org/2016/02/22/467210492/u-s-navy-brings-back-navigation-by-the-stars-for-officers https://www.npr.org/2016/02/22/467210492/u-s-navy-brings-bac... This (1960s?) US government produced (45m) video video gives a pretty good overview: * https://www.youtube.com/watch?v=UV1V9-nnaAs https://www.youtube.com/watch?v=UV1V9-nnaAs For those wanting to invest (substantially) more time, the two videos by "Tippecanoe Boats" are slightly rambling at times, but he does lay things out pretty well by the end of it (second is largely examples): * https://www.youtube.com/watch?v=-ARXW8InStY https://www.youtube.com/watch?v=-ARXW8InStY * https://www.youtube.com/watch?v=yu5R5mrrGB0 https://www.youtube.com/watch?v=yu5R5mrrGB0

PaulHoule 6y ago
This kind of weapon has a backup INS https://en.wikipedia.org/wiki/Joint_Direct_Attack_Munition https://en.wikipedia.org/wiki/Joint_Direct_Attack_Munition The INS doesn't need to be terribly high performance because it only needs to work for the time it takes a bomb to fall from the sky.
ardy42 6y ago
>> Military solutions will fall back to inertial, celestial or optical guidance, and people using GPS for navigation will at worst show up where they need to be somewhat later than planned. > The US Navy re-started celestial navigation a few years ago: Also, the SR-71, B-1, and B-2 and other aircraft have automated celestial navigation systems to provide corrections to their INSs: https://www.thedrive.com/the-war-zone/17207/sr-71s-r2-d2-could-be-the-key-to-winning-future-fights-in-gps-denied-environments https://www.thedrive.com/the-war-zone/17207/sr-71s-r2-d2-cou... Honestly, for military applications, GPS seems to be mainly useful for bad weather and providing navigation to the smallest units (and I'd think a small unit would be able fall back to a map and compass).
- mycall 6y ago
  ICBMs also use celestial navigation systems for same reasons.
matheusmoreira 6y ago
These military instructional videos are so great. They explain a lot of concepts very clearly. Nowadays videos seem to be more casual, with more focus on practice than understanding the theory.

aaron695 6y ago

Here's a good map/report on Russian in action GNSS spoofing - https://www.c4reports.org/aboveusonlystars https://www.c4reports.org/aboveusonlystars

henriklied 6y ago
Our news org did an investigation into this as well, and also demonstrated how a simple HackRF could be used to fool most modern GPS receivers: https://nrkbeta.no/2017/09/18/gps-freaking-out-maybe-youre-too-close-to-putin/ https://nrkbeta.no/2017/09/18/gps-freaking-out-maybe-youre-t...

throw0101a 6y ago

The OSNMA protocol discussed is based on Timed Efficient Stream Loss-Tolerant Authentication (TESLA): > This document introduces Timed Efficient Stream Loss-tolerant Authentication (TESLA). TESLA allows all receivers to check the integrity and authenticate the source of each packet in multicast or broadcast data streams. TESLA requires no trust between receivers, uses low-cost operations per packet at both sender and receiver, can tolerate any level of loss without retransmissions, and requires no per-receiver state at the sender. TESLA can protect receivers against denial of service attacks in certain circumstances. Each receiver must be loosely time-synchronized with the source in order to verify messages, but otherwise receivers do not have to send any messages. TESLA alone cannot support non-repudiation of the data source to third parties. * https://tools.ietf.org/html/rfc4082 https://tools.ietf.org/html/rfc4082 * A. Perrig, R. Canetti, J. Tygar and D. Song, “Efficient Authentication and Signing of Multicast Streams over Lossy Channels,” IEEE Symposium on Security and Privacy, pp. 56-73, May 2000.

oritsnile 6y ago
Part 1 also has a brief explanation of TESLA: https://berthub.eu/articles/posts/galileos-authentication-algorithm-part-1/ https://berthub.eu/articles/posts/galileos-authentication-al...

MayeulC 6y ago

Is there any cheap, open source alternative to GPS for navigation? I'm thinking of an electronic sextant (or star tracker) for rough position estimation, with a few km accuracy.

PaulHoule 6y ago
I'm somewhat annoyed at the accelerometer/gyroscopes that are available to the public, it is like they are all gimped to prevent you from building a weapons guidance system. For instance, the Wiimote doesn't have the dynamic range to handle the highest accelerations you can generate waving your arms, which makes it hard to use that kind of thing for athletic training.
- brandmeyer 6y ago
  > like they are all gimped to prevent you from building a weapons guidance system. You can buy RLGN off the shelf for a couple thousand bucks. You can buy gyrocompass-capable MEMS off of digi-key for dozens of dollars (Murata). Its just market forces and cost. You don't need super stable IMUs for game controllers.
- michaelt 6y ago
  They're not 'gimped' by any sort of conspiracy. It's just extremely hard to make a pure INS navigation system. Even if you could measure your acceleration accurate to 0.00001 m/s/s, because it gets integrated twice to give position even that tiny error is going to give you a position that's out by 100m after an hour. And measuring acceleration precise to 0.00001 m/s/s is a demanding task - like weighing a car so precisely you could count the pages of a book on the back seat. That level of precision simply isn't available in the affordable MEMS market segment.
  - generalizations 6y ago
    > They're not 'gimped' by any sort of conspiracy. It bothers me when people go around labeling random stuff conspiracies like that's a legit dismissal. If they were 'gimped', they wouldn't be the only hardware. GPS is still gimped and will shut off over a certain height and speed specifically to prevent use in missiles. And the precision used to be regulated, not by expense, but to give the military the superior tool.

PaulHoule 6y ago

Some people say that there isn't much open literature on GPS anti-spoofing, but there are many patents filed by the likes of Lockheed-Martin, Boeing, BAE, etc. I find the multiple antenna answers interesting. For instance, one of the easier attack scenarios against an airplane is to have a directional antenna on the ground. Because airplanes broadcast their GPS position via ADS-B, you could also know that you'd succeeded. In a case like that, however, the radio signal from the ground would be stronger than the signal from the sky and it would be obvious what was going on, unless the attacker managed to get the power level just right. With multiple receivers you also will see very different results with spoofing than with a real signal. For instance if you had a receiver at the front of the airplane and one at the back of the airplane, the time delay for all the fake satellites would be the same (they all come from the same place) whereas the time delays (e.g. position) would be noticeably different from real sats.

londons_explore 6y ago
if an attacker knows that the plane has two antennas, and their exact locations, he/she can generate the "correct" signals at each plane antenna with two attacker antennas and lots of math. Getting their exact locations is simplified by the fact the location is being transmitted by ADS-B...
- knorker 6y ago
  Nice. That sounds tricky though. You need to be very precise. The attacker needs to provide different received signal to two antennas meters (at most!) apart, at a distance of maybe kilometers, on a moving target. That's a hair thin angle. The aircraft just needs to TDoA "did it come from above or below (assuming an attack from the ground). That's 180 degrees. Or am I missing something? Unless I am, this doesn't sound feasible to me.
  - PaulHoule 6y ago
    The equipment I have pictured in my mind is one of the two-axis trackers that radio hams use to track satellites with a long but narrow Yagi-Uda antenna. (If that can handle the bandwidth) These tend to move in jerks and will get in real trouble if you try to move them over the zenith, but they do a great job with LEO satellites and would do OK to uplink one signal to an airplane. You might be able to hit two receivers if you had a phased array antenna like the Starlink antenna but bigger, but now it isn't a simple hacking project anymore. BTW, don't try it. There are certain things like aviation and nuclear power that "Posse Comitatus" doesn't apply to and you could find yourself looking down the barrel of an M4 carbine and getting frog-talk from the USMC much quicker than you'd expect.

knorker 6y ago

So if I understand this right the signature is 32bits times 24 times 15 per full frame? After the fact, why exactly can I not precalculate my spoofed data stream? I just need to spend: 3*2^31*24*15*(spoof_seconds/30) ops (on average)? (assuming 3 streams for position data) Sure, not cheap, but hardly hard even for even a hobbyist. So ~2^45 ops to spoof 10 minutes of data? That's doable. Is my math off? I have more faith in the direction finding aspects. Here's from the article, an understatement of the year: > To beat these simple tricks, a spoofer will need to have multiple transmitters that actually show the same parallax as the actual satellites. However, you can only do this by placing your transmitters next to the satellites - in space. This raises the bar significantly.

MauranKilom 6y ago
The HN formatting is eating your * signs...
- knorker 6y ago
  Thanks. Fixed.

knorker 6y ago

Another question for knowledgeable people here: Can I not just run two GNSS receivers 10 meters apart (on a ship), and if they report as having the same position, then I know someone is spoofing? It should be really hard to beamform the spoofing to spoof two different locations, at least from a distance because of the precise angles needed. Yes, this assumes that my real GNSS signal is good enough that they normally are 10M apart.

PaulHoule 6y ago
There's a patent on that one.
jaywalk 6y ago
The article addresses that scenario, and confirms that using multiple receivers would allow you to reliably detect spoofing. They'd report the same location but a slightly different time.
- knorker 6y ago
  Yes, but such countermeasures (like the ones mentioned in the article) are about one receiver doing fancy RF tricks. What I'm proposing is something entirely done by two cheap and existing receivers today, by the end user.