Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [P1619-2] Some text for the Introduction of 1619.2

Hi Shai,

thanks for writing this up, comments inline:

On Oct 13, 2008, at 9:09 AM, Shai Halevi wrote:


I put together a few paragraphs for the introduction of 1619.2. This
text explains the relations between .2 and .0 (and in particular also
the differences). It is, however, quite "dry" and technical, so it
would be nice if people can propose 1-2 more sentences. Maybe something 
about how the difference between .2 and .0 is expressed in potential
applications?

-- Shai


Introduction:

The purpose of this standard, similarly to IEEE-1619-2007, is to
describe a method of encryption for data stored in sector-based devices, 
where the threat model includes possible access to stored data by the
adversary. As in IEEE-1619-2007, this standard specifies
length-preserving encryption transforms to be applied to the plaintext
sector before storing it on the storage media.

Differently from IEEE-1619-2007,
I suggest replacing the above with "This standard improves on IEEE-1619-2007; " 


the encryption transforms that are
specified in this standard are "wide block encryption". This means that 
they act on the whole sector at once, where


I suggest replacing "where" with "and".


each bit on the input
plaintext influences every bit of the output ciphertext (and vise- versa for decryption). In particular, this standard specifies the EME2-AES and 
the XCB-AES wide-block encryption transforms.

Wide-block encryption can provide better protection than the
narrow-block encryption from IEEE-1619-2007 against attacks that involve 
traffic analysis and/or manipulations of ciphertext on the raw storage
media.
I believe that a stronger statement than this is warranted, perhaps "Wide block encryption better hides plaintext statistics, and provides better protection ..." 

best,

David