Re: AES and dynamic table generation

From: Mok-Kong Shen (mok-kong.shen_at_t-online.de)
Date: 01/08/04 

Date: Thu, 08 Jan 2004 19:10:43 +0100

Robert wrote:

> "Simon G Best" <s.g.best@btopenworld.com> wrote in message
>> I /think/ Robert means the generation of the tables used in optimised
>> implementations (you know, where the substitution and diffusion steps
>> are combined).
>>
>> If so, I'd imagine one use might be so that the tables don't have to be
>> stored permanently, with only the code to generate them being stored
>> instead.
 
> Thanks. You are right. Although my question is - is there any benefit in
> terms of increased security of the algorithm. From the discussions here, it
> seems like there is none (although there might be some small resource
> benefit).

As pointed out by Foo Bar, the algorithm and the
result of encryption are the same. Hence there is no
difference in security at all. The memory space for
the tables is also the same. On the contrary, there
is some additional computation cost for the dynamic
generation. On the other hand, one could in my humble
view more conveniently check the correctness of the
tables via examining the code parts that generate
them. (Opinions might differ on this.) My AES
implementation (see my web page) chooses to generate
the tables dynamically.

M. K. Shen
----------------------------------
http://home.t-online.de/home/mok-kong.shen

Relevant Pages

  • 3DES between .NET and VB
    ... I have recently implemented the 3DES algorithm using .NET. ... implementations I have found in VB 6 do not work correctly with this ... I imagine it is due to different string bases or some ... Joshua Ellul ...
    (microsoft.public.dotnet.framework)
  • 3DES between .NET and VB
    ... I have recently implemented the 3DES algorithm using .NET. ... implementations I have found in VB 6 do not work correctly with this ... I imagine it is due to different string bases or some ... Joshua Ellul ...
    (microsoft.public.dotnet.framework)
  • Re: AES and dynamic table generation
    ... terms of increased security of the algorithm. ... I'd imagine one use might be so that the tables don't have to be ... Simon ...
    (sci.crypt)
  • Re: Query about optimization
    ... implementations of N-bit full adder, one with simple statement like C ... the entity 'boundaries' at that point no longer ... from the top level code what the best algorithm to implement is....and then ... optomize the logic for that. ...
    (comp.lang.vhdl)
  • Re: Inverse error function (erf)?
    ... > algorithm and lists implementations in many languages (including bc ... "Lower tail quantile for standard normal distribution function. ... This function returns an approximation of the inverse cumulative ... The algorithm uses a minimax approximation by rational functions ...
    (comp.lang.lisp)