Re: How strong would randomizing data be ?

From: Joseph Ashwood (ashwood_at_msn.com)
Date: 05/03/05 

Date: Tue, 03 May 2005 08:44:14 GMT

"Skybuck Flying" <nospam@hotmail.com> wrote in message
news:d570at$128$1@news4.zwoll1.ov.home.nl...
> The question is if there is such a thing as true randomness ;)
>
> In the end everything happens because of some sort of system ;)

This is actually a somewhat open area of debate, no one has successfully
proven that entropy exists, but at this point the theories suggesting it
does not do not appear accurate.

As to the strength of your "design" it is almost certainly flawed, and you
haven't even given details. The reason is rather simple, you need to choose
cryptographically strong pRNGs and these are in short supply unless you use
a good block cipher in CTR mode. Either way you've over designed and under
engineered, especially in security.

As others have noted you have basically reinvented the stream cipher, albeit
in a rather convoluted way. You really don't need all the layers you have
and your reasons behind creating the design can be easily dealt with
otherways, the short problem is dealt with through padding. That you are
using ECB is easily addressed by switching modes, etc.
                Joe

Relevant Pages

  • Re: Simple block cypher for 8-bit microcontrollers
    ... Well, block cipher design and validation is very expensive, if you ... one round of the algorithm. ... This would be a kind of hybrid differential/slide attack. ...
    (sci.crypt)
  • Re: looking for research partner [block cipher design]
    ... > I have a design for a block cipher that is intended to be used in CTR ... > clocks per byte ratio [roughly 1.0625 cycles per byte]. ... After the first eight rounds the pipeline is filled ...
    (sci.crypt)
  • Re: help with block cipher design principles?
    ... >> Am a Research student and need some help with basics of block cipher ... >> design principles. ... I am a research student. ...
    (sci.crypt)
  • Re: Pin generation algorithm question
    ... >>The alternative is to design a block cipher with a peculiar block ... >>good block cipher is not especially easy. ... > no reason to design your own block cipher; ...
    (sci.crypt)
  • Re: ECB+CTR Mode?
    ... Your design requires both encrypt and decrypt algorithms. ... patterns in the plaintext, it can leak information about those patterns. ... CTR mode has the same shortcoming as all stream ciphers: ...
    (sci.crypt)