Welcome Guest [Log In] [Register]
Welcome to Crypto. We hope you enjoy your visit.


You're currently viewing our forum as a guest. This means you are limited to certain areas of the board and there are some features you can't use. If you join our community, you'll be able to access member-only sections, and use many member-only features such as customizing your profile, sending personal messages, and voting in polls. Registration is simple, fast, and completely free.


Join our community!


If you're already a member please log in to your account to access all of our features:

Username:   Password:
Add Reply
Base Translations Ciphers; Using Multiple Bases for Encryption
Topic Started: Jan 18 2014, 04:53 AM (3,378 Views)
mok-kong shen
NSA worthy
[ *  *  *  *  *  * ]
WTShaw
Mar 9 2014, 05:28 AM
You are using two group sizes. You must choose one or the other, or simply follow the design of the algorithm. Your choice of group size determines whether the leading zeros, if any, are assigned the zero place character from the present character set permutation.
The choice of the group size isn't the matter but the issue is "which" is the "first" character of the alphabet in relation to the first character (or characters) of the given specific plaintext!!! (I am very much surprised that you still don't capture this, although you are even the originator of the idea of the present base transformtion scheme.) If that's "X", then the given plaintext can't begin with "X". I have given an example in my Post#57. Please re-read it. If you still don't agree, then I "challenge" you to do the following: Choose an arbitrary group size, e.g. 6, and choose a plaintext of length 6 or less, i.e. there is only one "single" group and the group size becomes clearly irrelevant. Choose the plaintext such that it begins with the "first" character of the alphabet that you choose. Finally compute the number according to base conversion (in the "same" way that you did in your Post#52!) and subsequently decode the number back to plaintext and present your computation result with details with a post.
Edited by mok-kong shen, Mar 9 2014, 11:19 AM.
Offline Profile Quote Post Goto Top
 
mok-kong shen
NSA worthy
[ *  *  *  *  *  * ]
@WTshaw: To be extremely clear, I should additionally remark that in the examples in NUMCODING there is no concept of "groups" at all. Every plaintext string input to the function convtextnum() is treated as one single piece, resulting thus commonly in a relatively huge number. (If such huge integers are undesirable, the user has to divide his text into several pieces so as to get smaller numbers. For each such pieces, however, the constraint I repeatedly mentioned holds.)
Edited by mok-kong shen, Mar 9 2014, 06:12 PM.
Offline Profile Quote Post Goto Top
 
WTShaw
Advanced Member
[ *  *  * ]
@mok-kong-she: It would seem that your source of rules for me are inconsistent with my self-defined limitations. You miss the point in that I am implementing particular transform functions as I have pointed out in a discussion elsewhere. Such are closely and consistently defined. It also seems that you have either decided to change such rules governing these formulas for your own justification, applied rules and ideas for entirely different purposes, or have tried to perpetuate crippling and inadequate understandings, or stated for other reasons that you might wish to explain if you possibly ever understand what I have made work.

The constraints that some would artificially impose on cryptography for devious purposes of their own are often not founded functional realities and are therefore the essence of a some delusion. Be it as it is, some concepts cannot be falsely governed by inadequate precepts. And, hopefully, you have not partaken of the official Kool-Aid. Change for growth can be hard but necessary.

I deal with a series of numbers as Transitional Representations of cryptographic group but those numbers can usually very as to numbers of digits not defined by the transfer function in play. If the result is supposed to be in digits, leading zeros would be allowed to complete any group. You seem to believe that isolated groups can not exist in a group cipher, rather a strange non sequitur in itself.
Offline Profile Quote Post Goto Top
 
mok-kong shen
NSA worthy
[ *  *  *  *  *  * ]
@WTShaw: My best response to your last post is to "repeat" my challenge stated in my Post#91:

I have given an example in my Post#57. Please re-read it. If you still don't agree, then I "challenge" you to do the following: Choose an arbitrary group size, e.g. 6, and choose a plaintext of length 6 or less, i.e. there is only one "single" group and the group size becomes clearly irrelevant. Choose the plaintext such that it begins with the "first" character of the alphabet that you choose. Finally compute the number according to base conversion (in the "same" way that you did in your Post#52!) and subsequently decode the number back to plaintext and present your computation result with details with a post.

So please take up that challenge and don't reply that you currently have privately too much other stuffs to do (like you once did in the past in this thread in response to in fact the "same" challenge -- -- In Post#69 I challenged you, you replied with Post#70, saying you'll return to that issue. What have you done with that challenge till now???)!!!!!
Edited by mok-kong shen, Mar 11 2014, 01:53 PM.
Offline Profile Quote Post Goto Top
 
WTShaw
Advanced Member
[ *  *  * ]
I've been working to elevate more of the many associated algorithms to incorporate a standardized character set. Also, instead of using characters directly from the extended set, I am using character numbers in the source codes instead. Considering the multitude of candidates for updating, I work on a few at a time, right now closing on Pt's of 25 and 26 characters. (25 will not be useful for encrypted source codes since for example brackets and braces are not represented.)

At the other end of the spectrum, Pt's of 100 characters use some otherwise optional extended characters routinely merely to pad the total to 100. However, I will systematically use the same extended character for spaces between other characters and another for line feeds. Just because I did it one way does not suggest that different codings for packing the many keyboard characters into various Pt sizes are wrong as those would be suggested customized variations of such.

When I started toward implementations of Pt bases of 25 through 100, I went simple to have working examples at all levels. Later, codings were added. Now, as I bring relative algorithms up to my new standards, I am sure to add many additional algorithms as I comb through the possibilities. I'm having fun of course in my efforts but even as I may comment here and there on various topics in The Crypto Forum, I am generally for reasons and necessities of my own not here just to chase all rabbits that might have escaped within these environs.
Edited by WTShaw, Mar 31 2014, 01:30 AM.
Offline Profile Quote Post Goto Top
 
mok-kong shen
NSA worthy
[ *  *  *  *  *  * ]
WTShaw
Mar 31 2014, 01:27 AM
..... but even as I may comment here and there on various topics in The Crypto Forum, I am generally for reasons and necessities of my own not here just to chase all rabbits that might have escaped within these environs.
That's completely understandable. But could you "at least" respond to my last post which (importantly) contained a (many times repeated) challenge to you concerning a very essential point of our debate up till now???
Offline Profile Quote Post Goto Top
 
osric
Super member
[ *  *  *  * ]
@WTShaw

Using your base translation ideas I have developed a prototype dynamic concept. Below is a brief description. Full details at My WebSite with an interactive enciphering/deciphering program you can run in your browser.

I trust that I have not misrepresented any aspect of your work, or done it inadequate justice, but if you feel otherwise please signify.

The user inputs four things:
1. plaintext using any character up to asc 150,
2. a key,
3. the base to be used (max 150),
4. the size of enciphering block (max 90).

A substitution table is formed from the key. This is used to encipher the first block of plaintext, and the resultant intermediate ciphertext is expressed as digits. These are then combined at base 152 to form an integer, which finally is decomposed to the digits of the input base to form the final ciphertext for that block.

Each succeeding plaintext block is enciphered in the same way, after shuffling the substitution table in a way determined by the input key.

Deciphering is the reverse procedure.

There are several features of the base translation system that I like. Each plaintext character contributes to the ciphertext, so that changing a single letter in a block completely changes the ciphertext for that block. This makes attacks like hillclimbing impossible and emasculates a probable word approach. A statistical analysis is meaningless for the same reason.

Because the substitution is unknown to an attacker, he has no way to reverse engineer the enciphering process other than following a brute force approach. This will be fruitless because the key can be up to 150 characters long and each character can be selected from 120 different choices.

Making the cipher dynamic means that repetition of plaintext in different blocks becomes invisible to the cryptanalyst and robs him of a further tool.
Offline Profile Quote Post Goto Top
 
mok-kong shen
NSA worthy
[ *  *  *  *  *  * ]
@osric: So you are doing a multiple encrpytion (superencipherment) consisting of a (dynamic) substitution and a base conversion such that the ciphertext is in numbers instead of in alphabetical characters. That combination is new but dynamic substitutions, e.g. constant modification of the table of polyalphabetical substituton during encryption processing, isn't a novel concept IMHO.
Offline Profile Quote Post Goto Top
 
osric
Super member
[ *  *  *  * ]
Members of the ACA have been solving a simple cipher based on dynamic substitution since 1947, when such a cipher was first introduced in 'The Cryptogram'. In my own case I have been solving such ciphers since I joined the ACA in 2000.

Chaocipher, invented circa 1919, was perhaps the first cipher that used dynamic substitution. I spent a lot of time (too much!) in this Forum on Chaocipher.

So I am well aware that the principle of dynamic substitution is by no means novel in 2014.

Applying the principle to WT Shaw's Base Translation ciphers seems to me no more than natural development. In the sense that it hasn't previously been done by the author, to my knowledge, it creates a novel progression. What is of much more interest to me is that it creates a pretty secure cipher.


Offline Profile Quote Post Goto Top
 
1 user reading this topic (1 Guest and 0 Anonymous)
« Previous Topic · General · Next Topic »
Add Reply