Have I come up with a new unknown unb...

Have I come up with a new unknown unbreakable cipher method ?....

Posted in the Cryptography Forum

Since: May 11

Malmö, Sweden

#1 Apr 18, 2013
I think I have come up with a new quite simple cipher method that is unbreakable. It's some kind of Steganography method. It is similar to the one-time pad, but doesn't use modulation. Have someone seen this type of cipher before ? In that case, where ?

Anagram substitution:


Take a message to hide. Then take a text on another topic that include more than the same amount of each letter as the message to hide. Then find out which letters are "left-overs" from the text in relation to the message to hide. With these leftover letters, create a second text on some topic. Lets call the text A, and the left-over text B.


Subtract each letter from text A and also from text B, if that letter exist in both texts at each step.


Have someone seen this code before ? A decoder would need to know exactly which two sets of text to use.

Since: May 13


#2 May 31, 2013
It's clever, but it largely defeats the purpose of using a cipher code in the first place. The whole idea of pencil-and-paper cryptography is that if you have a long message which you wish to keep secret and the text in which the message is written is too large to easily hide, you encrypt the text with some manner of cipher method, using a much shorter key to encrypt/decrypt the text. That way, instead of hiding an entire book or other large document from an adversary, you only have to hide a small slip of paper from the adversary, and it doesn't matter if they find the book or other document. To do what you are suggesting, you would need two cipher keys to encrypt a single document, and each cipher key would probably have to be significantly bigger than the encrypted text itself. This means that successfully hiding the cipher keys to keep the document secret would be even more difficult than simply writing the document in plaintext and then hiding the document. It does sound like a fairly secure cipher. However, it would probably be easier and just as effective to simply use a one-time-pad. If you wanted to make it even more secure, you could use a running key cipher, which is to say a cipher using an existing text such as a novel, as the cipher key, and then use a one-time-pad on top of it. That way, someone would still have a very difficult time decrypting the message unless they were to find both keys, and neither of the keys would have to be any longer than the original message itself. That would be easier to manage than the method which you have described, and probably even more secure. You see, on the one hand, a one-time-pad is utterly impossible to decrypt without the key, but the key is very large and difficult to hide. On the other hand, a running-key cipher is possible, albeit difficult, to decrypt without the key, but the key is relatively easy to keep secret, i.e. when the inquisitor comes to your house, accuses your of heresy, and inquires about that botanical encyclopedia you used as a running-key to encrypt secret messages to other heretics, you can tell him "Oh no Sir, it's not a cipher key. It's just an encyclopedia." If you put a one-time-pad and a running key cipher together, it would certainly be easier to manage than the method which you described. I can't really imagine how someone would go about cracking your code without access to Text-B, the text containing the letters which are to be removed from Text-A in order to to produce the actual message. However, if Text-A was itself and intelligible text, such a a novel or encyclopedia, it would then be possible, and perhaps even very easy, for someone with access to Text-B to simply guess what Text-A was, simply by searching for texts containing most of the same letters as Text-B. For instance, if the message was "attackatdawn", and Text-A was the US constitution, Text-B would probably read something like this: "WETHEPEOPLEOFTHEUNITEDST [AT]ESINORDERTOFORMAMOREPERFEC [T]UNIONEST[A]BLISHJUSTI[C]EIN SUREDOMESTICTRANQUILITY ..." and so on. Already, anyone could easily guess what Text-A was. On the other hand, if Text-A was not an intelligible document but was instead just a random series of letters and numbers, it would be very obvious to anyone who found it that it might be the cipher key, and because Text-A AND Text-B both have to be bigger than the original message, it would be even harder to keep them hidden than to simply hide the original message. Thus not only would a one-time-pad together with a running key cipher be much easier to manage, it would also be far more secure as well. I don't mean to bust your bubble. Honestly, it is a very clever idea, and I can't imagine how someone could possibly crack your code without access to Text-B, but I think your method is far too cumbersome to be practical and the method I just described, of using a one-time-pad together with a running key, would be much easier and more secure.

Since: May 13


#3 May 31, 2013
While we're at it, how about examining my new cipher technique here:


My technique is designed not only to create as strong a cipher as possible, but to do it using as small and easily manageable a key as possible.

Tell me when this thread is updated:

Subscribe Now Add to my Tracker

Add your comments below

Characters left: 4000

Please note by submitting this form you acknowledge that you have read the Terms of Service and the comment you are posting is in compliance with such terms. Be polite. Inappropriate posts may be removed by the moderator. Send us your feedback.

Cryptography Discussions

Title Updated Last By Comments
Generic Substitution Cipher Help (I think) Aug 12 Damein 1
Cube26A Cipher Aug 9 pvial00 1
Galaxy Cipher Instrument open sourced. Jul '17 Russell Lee 2
Simple cipher I made. Jul '17 Mysteryman 1
HELP! I need help decrypting this! Jun '17 Anon 1
Who can decode this cypher and how long did it ... Jun '17 bhouseman 1
Scroll Cipher Instrument-open sourced. Jun '17 Russell Lee 1
More from around the web