info
Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Why Vector Ciphertext is Better.
10 views
Skip to first unread message
adacrypt
Oct 24, 2011, 5:49:50 AM10/24/11
to
If the domain of a keyset is made up of ‘n’ elements of non-repeating
vectors then there are n of the ‘i’ unit vectors, n of the ‘j’ unit
vectors and n of the ‘k’ unit vectors. These can be arranged to give
n x n x n or n^3 different larger composite vectors that may be used
as the
primary transformation of a single plaintext character in vector
cryptography.
Each of the elements of the larger keyset of n cubed vectors so formed
is totally disparate and independent of any other element in the same
set of n^3 so that there is no mathematical induction assumptions
whatever unlike scalar numbers that can be made of a long string of
ciphertext by a cryptanalyst.
For instance, given say the rth element he cannot deduce the (r + p)
element because individual elements when known, are only related by
irrelevant geometry in cryptography – there is no number theory as in
scalar data.
In scalar cryptography on the other hand there is always some
functional relationship between each and every one of the plaintext
that is passed on in a less transparent way to the ciphertext but
still there for the finding to a cryptanalyst. It’s transformation
into the separate elements of a string of ciphertext is only ‘n’ in
number compared to n^3 in vector cryptography. The elements of the
ciphertext string are always mathematically related to each other in
some way however difficult it may be to find this. The fact is that
numerical methods are there for
the finding of this relationship like differential analysis, linear
analysis, together with statistical and other direct inspection
methods like advanced Kasiski-Babbage metods that a cryptanalysts may
use.
The situation is fraught with what I call both data and system
‘structure’ that a cryptanalyst can exploit.
Pound for pound therefore vector data is more effective in capacity
and is much more secure against cryptanalysis.
A string of vector ciphertext is a string of natural (jump)
discontinuities in mathematics and it stymies all attempts at linear
analysis, differential analysis, cursory attacks like direct
observation of known plaintext attack, chosen plaintext, advanced
Kasiski- Babbage attack, totally unlike scalar numbers .
This resistance to cryptanalysis by vector ciphertext is a new
innovation in a highly specialised study that is frankly beyond the
ken of the writer but it is necessary for readers only to know that it
has a strong enough influence on cryptography that it is sufficient to
resist all forms of cryptanalysis (guardedly).
To a cryptanalyst who is geared to working on scalar ciphertext the
vector ciphertext environment is a no-go area of wildly disparate
(jump) discontinuities, the vector ciphertext string itself may be the
nearest thing to a “function” or “equation” of the vector ciphertext
space that anybody may ever be able to verbalise.
I think there’s the makings of very useful research study here and
perhaps a paper to publish the findings.
Warning.
– don’t go to the IACR with your paper – they have sold their souls
to Springer Verlag such that having worked your butt off preparing
your paper and overcome the distasteful intransigence of the
reviewers, and you are prepared to spend at least £5000 of your own
money on the vanity of hearing yourself speak on the rostrum of some
distant conference, for all that you’ll get a cut-price (not free mark
you – perish the thought!) book token from SV.
– isn’t life grand!
Instead of that, a well-crafted post here in sci crypt would be better
appreciated – anybody?
That’s my view – (don’t expect replies to rude or inane postal threads
from me - better still don't expext any).
- adacrypt
vectors then there are n of the ‘i’ unit vectors, n of the ‘j’ unit
vectors and n of the ‘k’ unit vectors. These can be arranged to give
n x n x n or n^3 different larger composite vectors that may be used
as the
primary transformation of a single plaintext character in vector
cryptography.
Each of the elements of the larger keyset of n cubed vectors so formed
is totally disparate and independent of any other element in the same
set of n^3 so that there is no mathematical induction assumptions
whatever unlike scalar numbers that can be made of a long string of
ciphertext by a cryptanalyst.
For instance, given say the rth element he cannot deduce the (r + p)
element because individual elements when known, are only related by
irrelevant geometry in cryptography – there is no number theory as in
scalar data.
In scalar cryptography on the other hand there is always some
functional relationship between each and every one of the plaintext
that is passed on in a less transparent way to the ciphertext but
still there for the finding to a cryptanalyst. It’s transformation
into the separate elements of a string of ciphertext is only ‘n’ in
number compared to n^3 in vector cryptography. The elements of the
ciphertext string are always mathematically related to each other in
some way however difficult it may be to find this. The fact is that
numerical methods are there for
the finding of this relationship like differential analysis, linear
analysis, together with statistical and other direct inspection
methods like advanced Kasiski-Babbage metods that a cryptanalysts may
use.
The situation is fraught with what I call both data and system
‘structure’ that a cryptanalyst can exploit.
Pound for pound therefore vector data is more effective in capacity
and is much more secure against cryptanalysis.
A string of vector ciphertext is a string of natural (jump)
discontinuities in mathematics and it stymies all attempts at linear
analysis, differential analysis, cursory attacks like direct
observation of known plaintext attack, chosen plaintext, advanced
Kasiski- Babbage attack, totally unlike scalar numbers .
This resistance to cryptanalysis by vector ciphertext is a new
innovation in a highly specialised study that is frankly beyond the
ken of the writer but it is necessary for readers only to know that it
has a strong enough influence on cryptography that it is sufficient to
resist all forms of cryptanalysis (guardedly).
To a cryptanalyst who is geared to working on scalar ciphertext the
vector ciphertext environment is a no-go area of wildly disparate
(jump) discontinuities, the vector ciphertext string itself may be the
nearest thing to a “function” or “equation” of the vector ciphertext
space that anybody may ever be able to verbalise.
I think there’s the makings of very useful research study here and
perhaps a paper to publish the findings.
Warning.
– don’t go to the IACR with your paper – they have sold their souls
to Springer Verlag such that having worked your butt off preparing
your paper and overcome the distasteful intransigence of the
reviewers, and you are prepared to spend at least £5000 of your own
money on the vanity of hearing yourself speak on the rostrum of some
distant conference, for all that you’ll get a cut-price (not free mark
you – perish the thought!) book token from SV.
– isn’t life grand!
Instead of that, a well-crafted post here in sci crypt would be better
appreciated – anybody?
That’s my view – (don’t expect replies to rude or inane postal threads
from me - better still don't expext any).
- adacrypt
adacrypt
Oct 24, 2011, 6:38:09 AM10/24/11
to
Globemaker
Oct 24, 2011, 7:25:14 AM10/24/11
to
On Oct 24, adacrypt <austin.oby...@hotmail.com> wrote:
> "Popular Cryptography Magazine" please copy. - adacrypt
Your teachings are being recorded and you will be paid $0.01 per
> "Popular Cryptography Magazine" please copy. - adacrypt
character for any mathematics or algorithmic documentation which is
posted on sci.crypt which meets the editorial standards of common
academia.
http://popcryfa10.blogspot.com/
Your words will appear on December 21, 2011 along with topics like:
The Hegemony CPU cryptographically secured instruction set
implementation in 24 nanometer silicon vault apparatus.
Alan Folmsbee, Publisher
adacrypt
Oct 24, 2011, 7:37:39 AM10/24/11
to
charity - also could you let me know of any past material that you
might want to publish - just say the word or indeed help yourself -
adacrypt
adacrypt
Oct 24, 2011, 8:14:15 AM10/24/11
to
until I have read your stuff) could it be possible to have that same
chip precharged with a large store of ready-to-run programs - I'm
talking in hundreds or even thousands of the same basic design.
Lemme explain.
You know my software contains programs Mark_0 , Mark_1, Mark_2'.....
After Mark_0 (the tutorial / diagnostic one) is complete creating
Mark_1 is easy - you simply delete the lines of source code that
output the commentary in Mark_0 and change the name to MarK_1.
Mark_2 is created simply by changing the name from Mark_1 to Mark_2
* If each subsequent change right up to Mark_1000 (or more) is
customised by setting the scrambling parameters to give a different
program setting each time then the operation of a multi faceted unique
piece of crypto hardware is created. The chip becomes the database
and the multitude of options of program setting enhances the databases
by the sheer proliferataion of options.- What a prospect!
I am interested to know what you think before throwing my inventions
in the public domain for nowt.
Contact me by private email if you want talk about this.
very serious - adacrypt
- adacrypt
Globemaker
Oct 24, 2011, 9:47:54 AM10/24/11
to
> I am interested to know what you think before throwing my inventions
> in the public domain for now.
>
> very serious - adacrypt
>
> - adacrypt
Dear adacrypt,
> very serious - adacrypt
>
> - adacrypt
The hardware CPU can include any software cryptographic algorithms in
its hardware logic. It is important to minimize that logic, so please
re-write your algorithm as several discrete functions like this :
$$$$$$$$$$$$$encryption$$$$$$$$$$$$$$$$$$$$$$$$$
input plaintext function
key input function from Alice
key schedule function to expand key to affect all plaintext bits
/* Each bit of the key must affect each bit of the plaintext! */
plaintext bits mixing function with all key bits
key output function so a small key can be sent to Bob
output ciphertext function
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
/* Explain the maximum file size that a plaintext can */
/* be for each key bit to affect each plaintext bit. */
/* In other words, if any one bit of the key changes, about */
/* 50% of the ciphertext bits will change. This is standard quality
control */
$$$$$$$$$$$$$decryption$$$$$$$$$$$$
Functions are subroutines that are called from a main() program. this
modular style helps people understand your stuff.
.
.
......
adacrypt
Oct 24, 2011, 10:28:43 AM10/24/11
to
the present concept of an external software that is customisable is
still the better way forward.
Also,
The concept of having multiple programs to hand as standard templates
or shells in the software that may each have a variety of inbuilt
settings of parameters that may still be tweaked a little as required
is still on - These are PROCEDURES in the Ada program designated
Mark_0, Mark_1, Mark_2 .... Mark_n.
The advantage here is that the common availability so to speak of 12
Ada language PACKAGES (large relative to a procedure) already created
at some expense of machine code that are needed to service one
PROCEDURE may also be used at no extra cost by 100's more procedures
that have not (proportionately) a lot of extra machine code.
This is just a design notion that still has to be justified some
acountable worthwhile benefit to users.
It is easily done by retrofit in any case so even if I open with say
Mark_0 , Mark_1 and Mark_2 I can always add ... up to Mark_100 or more
if needed.
Are you at all interest in marketing/promoting this crypto with me in
any form besides being embedded in a chip the way I proposed.
The constraints you outlined are a bit off-putting of that idea.
-adacrypt
Globemaker
Oct 24, 2011, 11:32:55 AM10/24/11
to
Dear adacrypt,
Yes, you have made a valid choice to withdraw your Vector Software
from being put into my hardware plans. The level of rigor and
verification for quality control is always much greater for hardware
manufacturing than that which is expected for software or mathematics.
I hope that you will continue to enjoy the devoted following which you
have cultivated on sci.crypt . God bless you and the hundreds of
acolytes and sycophants who dwell on your every word.
Yes, you have made a valid choice to withdraw your Vector Software
from being put into my hardware plans. The level of rigor and
verification for quality control is always much greater for hardware
manufacturing than that which is expected for software or mathematics.
I hope that you will continue to enjoy the devoted following which you
have cultivated on sci.crypt . God bless you and the hundreds of
acolytes and sycophants who dwell on your every word.
0 new messages