• Frank [he/him, he/him]@hexbear.net
    link
    fedilink
    English
    arrow-up
    11
    ·
    1 year ago

    inshallah-script

    I’ve heard the overwhelming focus on radar invisibility in a small number of radar bands doesn’t translate to the whole thing being radar invisible.

    I was also thinking about, with neural nets and large learning models, you could probably train a system to look for F-35 shaped anomolies in a radar network’s data the same way they train models to look for signs of cancer in a lung. One radar might not be able to see it reliably, but what about a series of networked radars using different wavelengths and methods, all networked together to create a single data stream for interpretation?

    • meth_dragon [none/use name]@hexbear.net
      link
      fedilink
      English
      arrow-up
      8
      ·
      1 year ago

      in all likelihood everyone’s had this technology for a while now. i lean a bit towards giving the f35 the benefit of the doubt though, and just assume that its signature is small and fuzzy enough for it to have a smaller confidence interval than other less stealthy aircraft

      same goes for submersible gliders equipped with passive sonar systems, just saturate an area with those and suddenly the ashbm kill chain is a lot more cloud cover resistant than nafo would have you believe

    • ☆ Yσɠƚԋσʂ ☆@lemmygrad.ml
      link
      fedilink
      arrow-up
      6
      arrow-down
      1
      ·
      1 year ago

      That’s my expectation as well, and on top of that, the jet necessarily produces emissions in a lot of different spectrums, such as heat and sound. So, I can’t see how you hide something like a jet from an integrated system that observes and integrates data across multiple spectrums. Back when computing power was low this was likely not practical to do, but it’s definitely something that would be done today.