Nothing is Hid – All will be revealed-2
adacrypt
old. In the original cipher the square is docked in the fourth
quadrant of the XY plane. I now decide to undock it from there and
translate it around the fourth quarter in equal distances from the X
axis and the Y axis. The upshot of doing this is to increment the
plaintext and the key values by an amount equal to X in every case so
that the equation of the square becomes,
[(PlainText +X) + (Key + X)] (Mod N) = Cipher text
I next look for values of N to use as moduli in a large-scale
experiment that has conditions attached to the properties of N.
Each individual N must divide the Sum [(Plaintext + X) +(Key + X)]
once and leave a residue that may be >= 0.
The effect of the latter is that the cipher text (as a residue modulo
N) requires compensating at decryption time by a single addition of at
least one N. The algorithm is contrived to require an addition of 2
N’s (may be more).
The effect also of incrementing the plaintext and the Key by the
amount X is a knock-on to the range of N’s that satisfy the conditions
– there is a linear relationship between X and the range of N so that
N is a burgeoning function of X.
There is a minimum starting value for X at 63.
The lower bound of the range of N is X + 127
The upper bound is 2(X + 32).
The values of Plaintext and Key are in the range 32 ..126.
The large set of moduli as N’s is collected and stored as a set of
random keys.
Note, the word ‘Key’ in the equation has been used verbatim simply to
keep in touch with the text in Applied Cryptography, it is the
‘alphanumeric’ key in a situation where the instantaneous modulus (N)
is also another key.
In practice the message is rounded to an exact multiple of the 95 key
set from ASCII so as to make that key random. The set of moduli as
keys are then used ad hoc to whatever the message length is so that
the encryption of a message is underpinned by two random keys that are
used once and have the same key-length as the message. Any message-
length is constrained only by the computer capacity to store positive
integers.
When the modus operandi is understood the crypto form of the algorithm
being expounded on http://www.scalarcryptography.co.uk can go over
entirely to presentation as a piece of modular arithmetic where it can
be expressed more easily for understanding by non-specialist readers.
The elements then become simply a set of integers being configured so
as to produce a desired result.
This cryptography uses the mutual database concept in which Alice and
Bob use exact copies of the same software and only harmless parameters
(as scramble and slice integers applicable to the hidden arrays) are
sent by email along with the cipher text.
The set of N’s as moduli are all different and effectively becomes the
periodic scale-set of an ever-changing scalar number-line that
contains the cipher text and the encryption key as points on it for
each encryption of a plaintext, hence the title “Scalable Key
Cryptography”.
The handbook can be dumped now, I went off that book years ago when I
read the chapter on randomness. In truth it is now useless as a
reference – it has become an anachronism by its very own dogma. – it
was a light reading reference at best but some of the material in it
is a joke – in passing, Bruce Schneier does not have a very high
opinion of sci crypt which he has said is political - have a look at
some of the postings in ‘sci crypt research’ that he promotes as an
alternative if you want some amusement - adacrypt
WTShaw
>... Bruce Schneier does not have a very high
> opinion of sci crypt which he has said is political - have a look at
> some of the postings in ‘sci crypt research’ that he promotes as an
> alternative if you want some amusement - adacrypt
Poor Bruce...he finds crypto kinda rough when he rides around on
binary sided wheels so much.
On the 32-126 set, Austin, very good as you are not a shadow boxer!
When you try to add in line/carriage returns through encrypted
characters, trouble looms as this is where the old Macintosh-UNIX-MS
lack of compatibility problem hits you. Everything is not solved, at
least not by the major contenders. The best solution may be to write
separate versions of a given program, but I'm playing with that
solution. It's all <32 business.
adacrypt
You've lost me a bit - do you mean that a decrypted message that was
encrypted in MS Windows will not auto carriage return in Mac when
decryted there ?
- I hope its as simple as that which I see as a mere management
problem - hope I am not being too simplistic in saying that - open to
any offers of help here - Austin