| 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: |
| Base Translations Ciphers; Using Multiple Bases for Encryption | |
|---|---|
| Tweet Topic Started: Jan 18 2014, 04:53 AM (3,379 Views) | |
| WTShaw | Jan 18 2014, 04:53 AM Post #1 |
|
Advanced Member
![]() ![]() ![]() ![]() ![]()
|
As threads go when starting discussions, we might go far afield. In a discussion on the usefulness of classical ciphers and new possibilities, I brought up a different and unique type. An older attempt to define them is My Webpagehere. I may need to repeat myself some from previous comments, but it is worth the struggle. So "Here we go again." Clean Base Conversion might be like 16 to 64, 27 to 81, etc., but Base Translation includes less exact conversions including when things don't quite fit exactly; almost is likely good enough in many cases. One fact is that there are many possible base translation ciphers, really unlimited except by imagination and making them work. They are also designed to invite whole new forms of attacks, just what our science should have. What I have done is mostly to cobble together suggestive primitives to encourage further study, and frustrate those that are looking for quick solutions of cipher-texts. Strange, new to you, processes are at work with base translation along with many old ideas. In general, it is "fun" to do them. I created many with javascript because of it's possibilities to work with most modern platforms and be open sourced. I appreciate input including criticisms while exploring the concepts and implementations involved. I too always have much to learn and thirst for that crypto liquor gladly. |
![]() |
|
| novice | Jan 18 2014, 08:44 AM Post #2 |
|
Super member
![]() ![]() ![]() ![]() ![]() ![]()
|
I find your Javascript programs confusing. Would you please gives us an example of enciphering by Base Translation. Take for example the following as plaintext and show us how you encipher it and what cipher text is produced: plaintext : "This is an interesting subject but I would find it helpful to be given a concrete example" |
![]() |
|
| Grip2000 | Jan 18 2014, 07:51 PM Post #3 |
|
no member
![]() ![]() ![]() ![]() ![]() ![]() ![]()
|
Hi WTShaw, have I interpreted it right? ![]() GRip |
![]() |
|
| WTShaw | Jan 18 2014, 09:18 PM Post #4 |
|
Advanced Member
![]() ![]() ![]() ![]() ![]()
|
The directions to go with bases 78 and 26 are: Pt26 and Ct78. Close is Pt77 and Ct26 which would also work. There are two of my algorithms that apply here, Providence and Chambal respectively. There are two sizes for each based on Pt group-size but let's take Providence4 for the example. Do the calculation first to see that 26^4 < 78^3, for a 96.3% efficiency (if I did it right). Mathematically, all possible Pt26 groups would fit whereas the other way around, 78 to 26, some possible base 78 groups would not convert cleanly to base 26, but Ct27, a sure thing. Think putting a hand into a glove of the right size... In encryption, the first goal is to use internal coding to cram most of the keyboard into base 26. It takes four subsets to do that with three selector characters which might appear differently on different computers. Not to worry, the actual keys are always indirectly changed based on word sequences and this level of actual key is never stored or pasted in. I'll use default unscrambled internal keys in the example. For reasons of strangeness, double quotes are not allowed in Pt26 encryption but two single quotes are. I'll omit the quotes and add a period: Pt: This is an interesting subject but I would find it helpful to be given a concrete example. Algorithm: http://cryptopopcorn.com/Crypto/js/Hybrids2/Providence4&8_Pome.html Default Internal Coded (Pt26) `abcdefghiklmnoprstuy~¢£±© (Pt26) Default (Ct78) 0123456789!@#$%^&*()abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ,./?'- (Ct78) Mt: E?g*& EC0T9 t&&ZH pkS*v ,KhGC 9ICFS FrK@Q )p4Sp DfL3) zsIuN %/).M 1T1Wu t(&(Z &fSov S,VwR g7U It's MetaText because this is before the final stream cipher step. Continuing through the stream cipher: Ct: C,h#Z dlNr, o8'Bw (9x2k DH(qt SfeuG ie*!S )@qe3 6Sv,? 77k6b OGVFE MyFaX 8-RSd MkY%H O4/w# 5M! Since I have worked with MacIntosh and Microsoft Computers with US options, I have no idea is something character-wise will happen with other country options. The program is semiautomatic for degliching, just use the buttons after each step to move previous results to the top data screen prior to the next encoding/decoding step, even through to the end when the original text should be produced. Complicated? Very, but this program does fly...at least here. Program is found at: http://cryptopopcorn.com/Crypto/js/Hybrids2/Providence4&8_Pome.html Scrambling and keyword changes are other subjects entirely as there are unlimited protocols that might be considered. We can break this down further as you want, or you can adapt it to more simplistic character sets like I stated out so long ago. |
![]() |
|
| novice | Jan 18 2014, 09:22 PM Post #5 |
|
Super member
![]() ![]() ![]() ![]() ![]() ![]()
|
@WTShaw Many thanks for the details which warrant some study before response. novice |
![]() |
|
| WTShaw | Jan 18 2014, 09:32 PM Post #6 |
|
Advanced Member
![]() ![]() ![]() ![]() ![]()
|
Of course I missed some of your question like the absent minded professor I am: 77x26 Chambal3-Pome26...Pt Base 77(3 char) <99.9%> Ct Base 26(4 char) http://cryptopopcorn.com/Crypto/js/Hybrids/Chambal3&6_Pome.html Pt with double quotes: "This is an interesting subject but I would find it helpful to be given a concrete example" (Pt77) qwertyuiopasdfghjklzxcvbnmABCDEFGH¢£±_0! 1$2@3&4%5*678#9^`-=+[]{}()<>;:'.,/?\| (Pt77) (Ct26) abcdefghijklmnopqrstuvwxyz (Ct26) Mt: culmw clckh ncnpn dxhnc wasam ssdvh dirjw dzymf lmlbr hdcgq hmarm auocg rxnev zdefy jbvht awble wgole dnldm mldul hmyol xbghf yehfi jbwkv ajlmi tptac ulm Ct: oeaca qoina achfd zxame exnuy ermle qrgis ydlua dvmpc jtpfr ykutg yaegt jzdzb ogzxt iaqos xqjiz kkmgf tdiej rucoq phkjq taaqv zfmwt fdmqf kmpke zrnfl tqw Encryption Coding uses three sets, two subsets. Much of the above reply still applies even as I got the cart before the horse, need coffee or tea I guess. |
![]() |
|
| novice | Jan 19 2014, 10:19 AM Post #7 |
|
Super member
![]() ![]() ![]() ![]() ![]() ![]()
|
@grip The enciphering system in your posting Nr 3 explains very nicely the method for encryption from base 77 to base 26 but may I suggest a slight modification to the top line of your table? It seems to me that it would more logically be26^4+1....26^3+1....26^2+1....26+1....1 and the lines below adjusted accordingly. Or have I missed something? |
![]() |
|
| Grip2000 | Jan 19 2014, 06:20 PM Post #8 |
|
no member
![]() ![]() ![]() ![]() ![]() ![]() ![]()
|
@novice You're absolutely right! I guess I was too tired that day. I've corrected it. Hope it is right now. I now have also attached the file. BR GRip |
![]() |
|
| Grip2000 | Jan 19 2014, 07:53 PM Post #9 |
|
no member
![]() ![]() ![]() ![]() ![]() ![]() ![]()
|
Even with random alphabets it seems to me as if the scheme would be "security by obscurity". instead of exponentiation ^ 3 this be better -> Page 9: http://www.rose-hulman.edu/~holden/Preprints/pohlig-hellman.pdf |
![]() |
|
| WTShaw | Jan 20 2014, 06:18 AM Post #10 |
|
Advanced Member
![]() ![]() ![]() ![]() ![]()
|
Some times better appearing source constructions may not be as efficient if done otherwise. Javascript sometimes needs remedial, lower level instructions or can get confused, as so it seems. "Security by obscurity" is a mere rationalization straight out of the 3-Letter Hymn Book. Confusion is operationally always a credible military tactic, and in out case a useful means of nibbling at strategic suggestions. Strange as it may seem to many, prime numbers are usually not always at the heart of useful number for crypto bases. There's lots going on in mathematics that seem to escape scrutiny. The use of meaningful headings for real charts used with examples of different bases is advised. For Base30 to Base13 look at: http://cryptopopcorn.com/Crypto/docs/Tables/SerenjeTables.html As far as tables go, I like this one best: http://cryptopopcorn.com/Crypto/docs/Tables/Base40Table.html It's easy to transpose the 8 digits for each group before solving for either Pt or Ct, but you might need a new space character in Ct. |
![]() |
|
| WTShaw | Jan 21 2014, 09:42 PM Post #11 |
|
Advanced Member
![]() ![]() ![]() ![]() ![]()
|
Novice said elsewhere: By introducing into the discussion the concept of using base change in cryptography, WTShaw has (in spite of much extraneous matter) touched on an interesting area. For example a simple combination of converting a plaintext into base 77 and then to base 26 followed by Vigenere encipherment might well be a headache to crack. The usual attack on Vigenere by statistical means would not work and the other way in through a probable word would be difficult. Brute forcing with a computer would have its limits if the key was long...[then, he gave up his insight.] Me: In keeping with strength as measured relative to an amount of data necessary to confirm a break, to is possible to have a message of greater length than many might expect. With a particular choice of text, algorithm, and selection of keys, that length surely wold vary but knowing that it might even be several hundred characters would be superior to the planned disasters of DES and AES where computer power alone is the difference. Viva John Gilmore et al. As the minister was said to the painter who diluted the paint to save his costs: "Repaint, repaint, and thin no more." Also remember that in cipher design it is an error to guarantee that the whole key is always used to the point of complete revelation as transitory nulls can otherwise muck up the barnyard. Mentioning the tendency of falling back to bad ideas, the other day I was speaking with 89 years old Clayton Pierce who was assigned the task of analysis of a personal cipher created/used by "Monte" in WWII. "Winston Churchill appointed Mountbatten head of Combined Operations Command on 27th October 1941. He launched a series of commando raids including the disastrous Dieppe Raid in August 1942. The decision by Churchill to promote Mountbatten to vice admiral, lieutenant general and air marshall ahead of older and more experienced men upset senior officers in the military establishment." Pierce said that the system although looking complicated was nothing more than a most common tableau cipher. History proves Monte's folly and the virtue of experience which he ultimately did gain the hard way. For learning purposes neophytes need to be know the quaint past of cryptography. But as shown, my Base Translation Hybrid Ciphers don't make that mistake. Just giving a cipher a different name does not mean it is not just a form of something else and/or weak...think, cogitate, prove. Edited by WTShaw, Jan 21 2014, 09:44 PM.
|
![]() |
|
| mok-kong shen | Jan 21 2014, 10:09 PM Post #12 |
|
NSA worthy
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]()
|
Is what fiziwig in the other thread and Grip2000 in this thread wrote about the "principal idea" of your software correct or wrong? You never seem to care to respond to others with concrete materials in a way such that one could continue fruitful debates with you IMHO. Above all, using math notations and numerical examples and, if required in essential points, a little bit of pseudo-code would greatly help the general readers to understand you stuff. |
![]() |
|
| WTShaw | Jan 22 2014, 07:53 AM Post #13 |
|
Advanced Member
![]() ![]() ![]() ![]() ![]()
|
Thanks for asking. I thought I had responded sufficiently but give me a brief example to show me what you want. The goal is to encrypt in such a way that requires much more to break it as compared to normal classical algorithms. Plus, a challenge should be serious without detailed hints and/or helps as in the real world. The principle idea is that multiple bases confound analysis by being a combined product of the strengths of the primitives involved. More ... a pseudocode example please, as my notes are in a form for me to follow using my own jargon. |
![]() |
|
| novice | Jan 22 2014, 09:40 AM Post #14 |
|
Super member
![]() ![]() ![]() ![]() ![]() ![]()
|
@WTShaw It's 'Monty' not 'Monte'. Some years ago Clayton Pearce sent me a copy of his (then) new book Neoclassical Cryptographic Systems II. In it he mentions the Mountbatten Field cipher, used in North Africa and parts of SE Asia during WW2. It was a Vigenere tableau with each column shifted by a different amount. I suppose that is what you are alluding to in your rather non-sequitur posting. I doubt that Monty personally used it -- he had other things to do than write field ciphers. With regard to your Base Translation Hybrid Ciphers, they appear to me to be a simple cipher that then has to be elaborated with several other layers to cover up the simplicity. I explained that in an earlier posting. The question then is: is it worth the effort? I think not. With less effort one can have a more secure cipher. |
![]() |
|
| WTShaw | Jan 22 2014, 10:09 AM Post #15 |
|
Advanced Member
![]() ![]() ![]() ![]() ![]()
|
I have a friend who spells it with a "y" but only heard the nickname for the Brit. I want you to know that I am fully aware of tableau ciphers, worked on them to see the variations maybe about 1960. More secure in what way, like DES? Shortly before he died that was is was part of the discussions I had with Cecil Phillips before the beginning of the AES process which ultimately also led to a crippled product. Not using all the key, or making it difficult to do so, helps. DES propaganda was a tipoff as I saw it in the files I reviewed. What is your suggestion? |
![]() |
|
| 1 user reading this topic (1 Guest and 0 Anonymous) | |
| Go to Next Page | |
| « Previous Topic · General · Next Topic » |





![]](http://z2.ifrm.com/static/1/pip_r.png)




but may I suggest a slight modification to the top line of your table? It seems to me that it would more logically be
Basis.xls (21.5 KB)
7:27 PM Jul 11