Re: Toaster to Generate Random Numbers

From: Barry Margolin (barmar@genuity.net)
Date: 01/10/03 

From: Barry Margolin <barmar@genuity.net>
Date: Fri, 10 Jan 2003 01:27:19 GMT

In article <v1s65ur2kvald7@corp.supernews.com>,
Guy Macon <. http://www.guymacon.com/resume.html .> wrote:
>Barry Margolin wrote:
>
>>Although the outsider has access to all the messages, he doesn't know which
>>messages you digested and in which order you scanned them to produce your
>>random seed. Also, there's quite a bit of server-specific data, such as
>>article numbers, and Path and Xref headers, so unless he's using the same
>>news server as you are he won't get the same results.
>
>In other word, you randomly picked a newsserver and randomly picked
>the articles to process, keeping those decisions secret. Have you
>really increased the entropy over that of the RNG you used to do
>the choosing?

I think so.

Suppose you have an RNG that chooses a random integer from 1 to 10. You
keep the past 10 days of newspapers, and use this number to select which of
them to digest to seed your random number sequence.

At first glance this doesn't seem to be any better -- you're just selecting
among 10 random number sequences, which you could just as easily do with
the original integer. But the difference is that the function changes
every day. If the simple RNG produces 1 today and 1 tomorrow, you'll get
different sequences.

I admit that I'm not a mathematician and I haven't done a detailed
analysis. The above example is really simple and has flaws (e.g. 1 today
is the same as 2 tomorrow), but I expect that it could be improved upon to
produce good results (e.g. when a newspaper is used, take it out of the set
so that it won't be used again). 

-- 
Barry Margolin, barmar@genuity.net
Genuity, Woburn, MA
*** DON'T SEND TECHNICAL QUESTIONS DIRECTLY TO ME, post them to newsgroups.
Please DON'T copy followups to me -- I'll assume it wasn't posted to the group.

Relevant Pages

  • Re: Toaster to Generate Random Numbers
    ... ]Suppose you have an RNG that chooses a random integer from 1 to 10. ... ]keep the past 10 days of newspapers, and use this number to select which of ... ]them to digest to seed your random number sequence. ... ]different sequences. ...
    (comp.security.misc)
  • Re: parallel random numbers generator
    ... simulation are alike, I guess, to needs of most of MC based ... simulations to be implemented on CUDA, ... to have these sequences manifesting low correlation between ... by another RNG). ...
    (sci.math.num-analysis)
  • Re: Finding independence seq. of Random numbers
    ... and each sequences has to be independent. ... You could use alternative types of RNG: ... However, using three different seeds will generally be independent, ... Because you need to restart sequences many cryptographic RNGs will not ...
    (comp.lang.java.programmer)
  • Re: Don Knuth - Uncompressable sequences
    ... "We can compress sequences that Stanford University professor Don ... Knuth here discusses a couple of heuristic tests to check the quality of a PseudoRNG, and what the above statement means is likely that there are sequences that pass those tests that can be still compressed. ... It isn't hard to generate PseudoRNGs that are easly compressible - namely just by one number, the seed of the RNG and the code for the RNG. ...
    (comp.compression)