Re: New Paper

From: Tom St Denis (tomstdenis_at_iahu.ca)
Date: 02/01/04 

Date: Sun, 01 Feb 2004 03:06:46 GMT

"Tom St Denis" <tomstdenis@iahu.ca> wrote in message
news:5uDSb.77785$fgk.58380@news01.bloor.is.net.cable.rogers.com...
> Title: The CSQUARE Transform
>
> Abstract: In this paper we show how to combine the design concepts of the
> SQUARE and CS block ciphers to produce a pseudo-random permutation CSQUARE
> suitable for use in block cipher and hash design with a very high
> multi-round trail weight. The new design inherits the hardware efficiency
> of the SQUARE linear transform pattern as well as the efficiency of the
fast
> pseudo-Hadamard transform over a finite field. We demonstrate the DMWT
hash
> function which makes use of our new results.
>
> URL: http://libtomcrypt.org/dmwt.pdf
>
> Note: It's a draft and work in progress. I appreciate all comments,
> critiques, etc...

I've updated the paper [same URL, the new copy will have todays date on the
bottom, January 31st...]. Mostly I've beefed up the discussion in the paper
and added references to SHARK. I've changed my DMWT demo hash from the
Davies-Meyers to the same style WHIRLPOOL uses ["Scheme 10" from a Crypto'93
paper...].

I've also decided to make the implementation of DMWT public domain on the
site [http://libtomcrypt.org/dmwt.c]. Note that I don't suggest people use
the hash. I'm only releasing the code for people to see a concrete
implementation of the design.

[And Yes, I've submitted the paper to eprint. I'll send the updated copy
when they publish the first copy...]

Tom

Relevant Pages

  • New Paper
    ... Title: The CSQUARE Transform ... In this paper we show how to combine the design concepts of the ... SQUARE and CS block ciphers to produce a pseudo-random permutation CSQUARE ...
    (sci.crypt)
  • Re: [PROPOSAL/PATCH] Fortuna PRNG in /dev/random
    ... was good not to rely *entirely* on the ahsh algorithms. ... The point is that the current random.c design DOES NOT ... RELY on the security of the hash function. ... plaintext that's not one of the two. ...
    (Linux-Kernel)
  • Re: Tenacity should be rewarded
    ... constructed tree, its a hash table and with hash tables, you either ... create a "perfect hash" design or you design the collision handling ... be individually located with the collision handling code. ... Arrrrgh, CUM the REVOLUTION KOMRADE, your assembler coding can be as ...
    (alt.lang.asm)
  • Re: Does this have a flaw in de-biasing an entropy stream?
    ... It takes the same time to build a bad circuit as it does to build a good circuit. ... The best time to stop this is in design time, ... the output to the user should be the output from a hash function ... If the hash function's output is n bits, feed raw data into it until by some estimate the amount of entropy in the raw data has reached n bits or so, then output the hash value, and start again with more raw data. ...
    (sci.crypt)
  • Re: expensive password hash
    ... >>phrase or he will not detect when he guesses the right pass phrase. ... I fail to see how this is done in your design. ... The hash over the buffef full with hashes is used to ... In this design the number of iterations or the time to waste is not ...
    (sci.crypt)