Re: Hash of item as IV for CBC mode?

David Wagner wrote:
Tim Smith wrote:
Are there any known problems with using a hash of an item as the IV for encrypting that item using a block cipher in CBC mode?

Yes, it violates semantic security: if I encrypt the same item twice, then
that will be visible from their ciphertexts. If the space of possible
items is not too large, then this is basically a simple substitution
cipher on items.
{snip}

Appending the time and date of the encryption to the start (or end) of
the message will produce a different hash value each time the message is
sent.

Andrew Swallow
.

Relevant Pages

  • To all who responded. [Re: Q: Vigenere]
    ... > Nope, can't see why it would, unless practice hitting the backspace key is ... My app, of course, removes those automatically before encrypting. ... I haven't actually compared the digraph distribution of English ... > Preserving word spacing in any form generally weakens the cipher. ...
    (sci.crypt)
  • Re: Ive seen things you people wouldnt believe...
    ... TrueCrypt, burn the resulting blob of randomness to DVD, then ... TrueCrypt is a program for encrypting entire hard-drives using AES ... with a random key, generates something that, if you use a good cipher, ...
    (rec.arts.sf.fandom)
  • Re: Hash of item as IV for CBC mode?
    ... encrypting that item using a block cipher in CBC mode? ... Yes, it violates semantic security: if I encrypt the same item twice, then ... Why are you encrypting a serial number that increments? ... authenticity/integrity. ...
    (sci.crypt)
  • Key generation for multiple blocks (block cipher theory)
    ... key, a set of subkeys is derived, which are used when ... encrypting a block. ... I propose a cipher which encrypts every subsequent block with ... supplied by an theoretically infinite key schedule. ...
    (sci.crypt)
  • Re: Hash of item as IV for CBC mode?
    ... as the IV for encrypting that item using a block cipher in CBC mode? ... he was encrypting items for long-term storage. ... As Dr. Wagner pointed out, this is akin to using the cipher in ECB mode and, while it's true that this can be thought of as a substitution cipher, it might be very difficult to construct the substitution table needed to break it. ...
    (sci.crypt)