In the Krack-attack paper, the section 6.3 says that a vulnerability on Android devices involves installing an all-zero encryption key when the message 3 is re-sent.

In his demonstration, he shows via Wireshark that the credentials of the user were sent in plaintext.

How can any algorithm result in a cipher equal to the plaintext if it's key is all-zero ? Am I missing something ?


If the key and nonce are identical then it is possible to recreate the counter mode key stream. WPA2 is using CCM, which uses counter (CTR) mode underneath. This basically generates a key stream which is XOR'ed with the plaintext message. If you have more messages then it becomes easy to retrieve messages by XOR'ing multiple streams together.

If the key is an all zero key then all the adversary has to do is to decrypt the ciphertext of course. Once you have the key then all the security is lost. However, the ciphertext will still not be the same as the plaintext, so what Wireshark is showing in the video must be the decrypted ciphertext.

In other words, the explanation is slightly over-simplified in my opinion.

  • $\begingroup$ "the explanation is slightly over-simplified in my opinion", that's what I expected to hear .. Thanks for the answer ! $\endgroup$ Jan 21 '18 at 13:33
  • 2
    $\begingroup$ Put differently, (symmetric) encryption with a known key is just an encoding. $\endgroup$
    – yyyyyyy
    Jan 21 '18 at 18:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.