• notabot@piefed.social
    link
    fedilink
    English
    arrow-up
    20
    ·
    2 days ago

    It seems odd to just transmit wildly false coordinates, as any receiver could easily be designed to detect it’s way out of theatre and swich to an alternative system or, at least, ignore the erroneous data. Sending subtly wrong data instead would seem like it would have a more reliable effect. If a missile think it’s constantly a mile too far north, for instance, it’s going to turn until it’s pointing south. Keep adjusting the signal, and you can probably herd it back to it came from.

    That they’re not doing that suggests that either there’s a technical reason for it, or the report is obfuscating details of the tactic, which would seem to be a wise plan.

    • SapphironZA@sh.itjust.works
      link
      fedilink
      arrow-up
      2
      arrow-down
      1
      ·
      5 hours ago

      I think from a technical perspective it’s less suspicious to trick the drone to think it is in Peru, compared to making it think it’s 10km off course.

      The way positioning systems work is they look at time signals from multiple satellites. By looking at when the signals arrive and the time encoded in the signal. If you can mess with the arrival time of the signal through jamming, to move it 10km, you need to mess with the signal of just one satellite and not the others. Then if the computer has extra signals to work with, it can detect and ignore the bad signal. That’s a very basic math operation that can be done by a very basic computer. 1960s level stuff.

      But if you can delay the time signal on all the sources by the same amount, the computer is non the wiser, because they are all corroborating each other. The effect is it places it on the opposite side of the globe. In order to prevent this form of spoofing, you need a higher level computer to interpret what the guidance computer is telling it. It needs memory to track where it is reported to be and where it is expected to be. It likely also needs additional optical or radar sensors to fall back on.

      This drives up cost and complexity.

      • notabot@piefed.social
        link
        fedilink
        English
        arrow-up
        2
        ·
        3 hours ago

        Detecting that the signal is false when it’s reporting the missile or drone is in Peru is a simple as noticing the coordinates are outside the expected operational area, so little more than two range checks, one for latitude, one for longitude. Determining that the signal is only a little off is much harder without also running inertial navigation to compare it against, and even then you have the issue of which to trust.

        GPS receivers have the ephemeris data of all the GPS satellites, so just delaying all of the signals again wouldn’t work, as the receiver would know those particular satellites aren’t above the horizon in Peru. I suspect that the way to spoof the signal location is to jam the normal signals (which are pretty low powered by the time they get to to receiver) and transmit faked signals for all satellites that the receiver is expecting. Doing that would let you feed it a more believable error.

        • SapphironZA@sh.itjust.works
          link
          fedilink
          arrow-up
          1
          ·
          1 hour ago

          Those satellites orbit the earth, so depending on how and how much you spoof the time signal, it may make complete sense according to the GPS signals for those satellites to be over Peru.

          Remember these military guidance kits are exceptionally basic, so as to be more reliable and less sensitive to basic forms of jamming. Which is why spoofing them becomes highly situational.

          Civilian GPS equipment is infinitely more sophisticated and has a lot more processing power available to it. Because it does not need to deal with an electronic warfare environment.

    • cynar@lemmy.world
      link
      fedilink
      English
      arrow-up
      6
      ·
      2 days ago

      My guess is that military systems can detect potential spoofing, even quite subtle, and abandon GPS for inertial guidance. If you’re subtle enough for it not to notice, it won’t throw it off far enough to be useful. This is still a win however, since that will hit accuracy hard still.

      I also suspect that cutting past valid, but incorrect signals is harder than the equivalent amount of interference (white noise). This would cut down on transmitter power requirements, making it cheaper and harder to target. The wildly inaccurate data (rather than subtly) would also help avoid accidentally screwing over friendly forces also trying to use GPS.