Cryptography-Digest Digest #944, Volume #10 Fri, 21 Jan 00 07:13:01 EST
Contents:
Re: RSA survey (The King)
Re: RSA survey (NFN NMI L.)
Re: What about the Satanic Seven??? ("Bayard Randel")
Re: What about the Satanic Seven??? (David A Molnar)
Re: Codebook URL (Arturo)
Re: LSFR ("r.e.s.")
Re: Transposition over ASCII-coded text ("Markus Eiber")
Re: What's with transposition? ("Markus Eiber")
----------------------------------------------------------------------------
Date: 21 Jan 2000 07:20:18 -0000
From: The King <[EMAIL PROTECTED]>
Subject: Re: RSA survey
If you think it is big enough it usually isn't <grin>.
Remember that all the really clever and gifted people in this world do
not allways talk, perhaps somewhere, someone has developed new and
better methods for breaking these keys.
Cryptographers, by the very nature of their calling should not be
hidebound and overlook this possibility <grin>.
Peace.
------------------------------
From: [EMAIL PROTECTED] (NFN NMI L.)
Subject: Re: RSA survey
Date: 21 Jan 2000 07:24:59 GMT
That Scott Nelson post seems symmetric. Bzzzzt!
<<I use big keys myself, because overkill does not cost extra>>
Rock on, man, rock on.
S. "There is no word in the English language for just the right amount of kill"
L.
------------------------------
From: "Bayard Randel" <[EMAIL PROTECTED]>
Subject: Re: What about the Satanic Seven???
Date: Fri, 21 Jan 2000 20:32:11 +1300
>
> Well, Microsoft did this really complex location check to
> prevent anyone outside the us to download the 128 bit Service
> packs for Windows NT;
>
> First by Ip, Then by checking the BROWSER, because you could
> trick the download wizard via altavista's babelfish...
> (How did i know that? ;o)
MS does a standard whois internic query to ensure that you are in US-DOM
when exporting strong crypto.
This is very silly of course, as anyone can use a public http proxy, or
shell account while outside of the US. I don't think they have considered
this in great depth.
Bayard Randel
Christchurch, New Zealand
------------------------------
From: David A Molnar <[EMAIL PROTECTED]>
Subject: Re: What about the Satanic Seven???
Date: 21 Jan 2000 08:01:39 GMT
Bayard Randel <[EMAIL PROTECTED]> wrote:
> MS does a standard whois internic query to ensure that you are in US-DOM
> when exporting strong crypto.
> This is very silly of course, as anyone can use a public http proxy, or
> shell account while outside of the US. I don't think they have considered
> this in great depth.
You're assuming that the reason they did the query was to stop the
export of strong crypto. Wheras it may have been the case that they didn't
care about export, but were taking the minimal steps necessary to comply
with U.S. export regulations. Until the new regs came out, there was a
provision in the previous set which said that it was OK to post crypto
code, as long as some checking was done to prevent foreigners from
downloading.
As I recall, the requirements for such checking were pretty minimal.
So no reason for MS to invest any time or effort at all in improving over
reverse lookup.
Now, of course, the requirements are gone! :-)
-David
------------------------------
From: [EMAIL PROTECTED]=NOSPAM (Arturo)
Subject: Re: Codebook URL
Date: Fri, 21 Jan 2000 08:08:23 GMT
On Thu, 20 Jan 2000 21:21:00 GMT, [EMAIL PROTECTED] (Mike Andrews)
wrote:
>I'm in the process of scanning in and making available a WW-II
>US Army training codebook. The URL is
> "http://mikea.ath.cx/codebook".
>It's incomplete, but I get a bit farther into the stack of
>pages every day.
>
>Also got some US Army training material on cryptanalysis
>and on construction of Signal Operating Instructions (SOIs),
>which I'll be putting up as time permits.
Great. And, if you want more manuals, may I point the
following url:
http://www.und.nodak.edu/org/crypto/crypto/army.field.manual/
(a US field manual). Enjoy.
------------------------------
From: "r.e.s." <[EMAIL PROTECTED]>
Subject: Re: LSFR
Date: Fri, 21 Jan 2000 01:59:38 -0800
"David Wagner" <[EMAIL PROTECTED]> wrote ...
[re 10-register base-10 LFSR]
: I think you need to look at the feedback polynomial
: mod 2 and mod 5. If both of those are primitive, I'd expect the
: period to be either (2^10 - 1) * (5^10 - 1), 2^10 - 1, 5^10 - 1, or 1.
:
: For example, if I initialize the register with all zeros, it stays all
: zeros. If I initialize the register with values that are all even,
: it stays this way. If I initialize the register with values that are
: all divisible by 5, this too stays the same. So (at best) there will
: be four cycles, one of each of the lengths mentioned above.
OK. In that case, since I'm interested mainly in the n-register LFSR
with x(i) = x(i-n) + x(i-n+1) (mod 10), where there are n initial values,
I have two main questions -- sorry to be so pedestrian about this:
1) Is the polynomial x^n + x^(n-1) + 1 primitive as required?
2) If so, is the proportion of possible initial-value vectors that
produce the maximum cycle length just (2^n - 1)(5^n - 1)/10^n?
And third, a more practical question, being optimistic about the first two:
3) Wouldn't non-linear post-processing effectively destroy vulnerability
to Berlekamp-Massey? (For example, collect a few rows of output digits,
n digits per row, then read them out by columnar transposition.)
Although these methods were used in the "VIC" cipher (with n=10) to key
a double transposition, couldn't they be used with larger n to get
something adequate as a stream cipher, still workable by hand?
(n registers would yield ~n/0.3 bits of entropy; also, if the above
questions have optimistic answers, it looks like there's only a 1/2^n
chance of missing the maximum cycle-length.)
--
r.e.s.
[EMAIL PROTECTED]
------------------------------
From: "Markus Eiber" <[EMAIL PROTECTED]>
Subject: Re: Transposition over ASCII-coded text
Date: Fri, 21 Jan 2000 12:18:29 +0100
--
Markus Eiber
Werner-Heisenberg-Weg 106
85579 Neubiberg
Germany
[EMAIL PROTECTED]
____________________________
"Gew�hnlich glaubt der Mensch, wenn er nur Worte h�rt,
es m�sse sich doch dabei auch was denken lassen."
Goethe
Mok-Kong Shen schrieb in Nachricht <[EMAIL PROTECTED]>...
>Markus Eiber wrote:
>>
>> I study electronic engineering having a focus on transmission units.
>> In digital transmission units there is often used a scrambler or an
>> interleaver in order to improve error correction for the transmitted
data.
>> In general those functions could be interpreted as a transposition of the
>> digital data. Let this data be for instance be ASCII-coded (8 bit) text
and
>> let us assume, that we don't know the key for transposition. If we decode
>> the scrambled bit stream we will note, that the result is a polygraphic
>> substitution (if the length of one transposition_block is not equal to 8
>> bit). Depending on this length the redundancy of the transmitted text
goes
>> down and with great length, it will be very difficult to break the
>> substitution (Hill-cipher!). Anyway, we are still in posession of the
>> digital data. My question is, if there is a way to break the cipher using
>> the digital bit stream (this means not to decode it) when only ciphertext
is
>> available?
>
>I know much too little about electronics, hence some questions
>of ignorance:
>
>(1) Does one use in 'normal' transmission a 'scrambler' to improve
> 'error correction'?
Yes. First of all the data gets scrambled and then redundancy for error
correction is added. During transmission there often occur burst errors
(several neighbouring bits get falsified) which cannot be corrected (common
codes can correct only up to two bits per codeword). So several codewords of
information get lost. One can avoid this by scrambling the data, because
with descrambling the falsified bits get distributed to a lot of codewords,
that is the quantitity of false bits per codeword gets reduced.
> I thought that various much-studied codes
> with capabilities of doing error-correction are used. But
> these are in my view not scrambling in the sense of encryption,
> since these codes are well-known (public).
>
>(2) If one uses a single error-correcting code, then one has a
> monoalphabetic substitution. Where does polyalphabetic
> substitution come from?
I never said something about polyalphabetic substitution! I mentioned, that
the composition of the transposition of ASCII-coded text and decoding the
ciphered bit stream results in a polygraphic substitution (which is
of course monoalphabetic)!
>
>(3) I am yet ignorant of any use of Hill's method in the context
> of transmission error-correction.
Maybe I didn't get it to the point because of may poor knowledge in English.
As I mentioned above, data gets scrambled which is equivalent to a
transposition. Basically the function to this transposition (scrambling) is
public, but who can prevent us from creating a secret key depending
function? So let us fix the point that we have ASCII-coded text which is
ciphered by a simple transposition with a great block length (e.g. > 500
bits).
I am looking for an efficient way to analyze the ciphered bit stream (is for
instance correlation useful?).
(Little note to Hill's method in this context: If one decodes the ciphered
bit stream he gets a monoalphabetic polygraphic substitution of the
plaintext, this is what Hill's method creates, too)
> Could you please give a
> literature reference?
>
>Thanks,
>
>M. K. Shen
------------------------------
From: "Markus Eiber" <[EMAIL PROTECTED]>
Subject: Re: What's with transposition?
Date: Fri, 21 Jan 2000 12:33:31 +0100
Douglas A. Gwyn schrieb in Nachricht <[EMAIL PROTECTED]>...
[..]
>Of course, the same kind of thing can be done in binary if one
>has a good idea of the plain-text "language", but it usually
>requires more formal mathematical methods for finding the most
>likely permutation.
These formal mathematical methods are exactly what I'am looking for. Can you
give me a hint or a literature reference for how one can cryptanalize the
transposition of a binary coded plaintext?
Thanks for your efforts!
Markus
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list (and sci.crypt) via:
Internet: [EMAIL PROTECTED]
End of Cryptography-Digest Digest
******************************
