1. Subscribersonhouse
    Fast and Curious
    slatington, pa, usa
    Joined
    28 Dec '04
    Moves
    52619
    19 Dec '13 15:27
    http://phys.org/news/2013-12-trio-rsa-encryption-keys-noise.html

    Now what, sports fans?
  2. Joined
    31 May '06
    Moves
    1795
    19 Dec '13 16:341 edit
    Originally posted by sonhouse
    http://phys.org/news/2013-12-trio-rsa-encryption-keys-noise.html

    Now what, sports fans?
    Hmmm.

    They are not actually breaking the encryption itself. They have created a
    novel method for extracting the key from the target machine without
    necessarily infecting that machine with a trojan or hacking into it.

    Which is obviously a security threat... But it's not a failing of the encryption
    system itself. It's always been possible to break encryption by hacking the
    machine with the keys on it. What this does is make THAT easier.
    It doesn't make it any easier to decode the message on route.


    I would suspect from the description that a one time pad encryption system
    would be invulnerable to this attack as the cypher changes with every bit.
    So as long (as ever) the key has a high enough degree of randomness and
    is kept safe this attack would fail.

    Of course that gets you back to wondering how you safely transport the key...
    Quantum encryption isn't ready yet.
  3. Subscribersonhouse
    Fast and Curious
    slatington, pa, usa
    Joined
    28 Dec '04
    Moves
    52619
    19 Dec '13 18:27
    Originally posted by googlefudge
    Hmmm.

    They are not actually breaking the encryption itself. They have created a
    novel method for extracting the key from the target machine without
    necessarily infecting that machine with a trojan or hacking into it.

    Which is obviously a security threat... But it's not a failing of the encryption
    system itself. It's always been possible to br ...[text shortened]... you back to wondering how you safely transport the key...
    Quantum encryption isn't ready yet.
    One method that will probably work is when spintronics gets going big enough to rival modern CPU's and memory. The gist of that is the energy requirements are less than one tenth of one percent of normal CPU's and therefore would be that much more quiet and therefore that much harder to decode directly.