We're going to China

My wife and I leaving for China in the very near future. We're both really excited, I've never been to the eastern hemisphere before and I'm looking forward to experiencing a different culture. The trip will also give Vanessa plenty of opportunities to practice her Chinese.

I'll have a vast array of experiences to share with you, gentle reader, upon our return. So for now I'll leave you with some half finished thoughts. I started no less than three blog posts over the past two weeks, but they are all sitting as unfinished drafts. Here's a brief summary of one of the things I started to talk about. (In the end it turned out to be not quite so brief.)

Cellular Automata
I've been looking into Cellular Automata recently as a source for complex data interactions. There has been some discussion among cryptographers about the use of cellular automata as possible pseudo random number generators. The best known example of a cellular automata is John Conway's Game of Life and I think it illustrates the kind of complex interactions which can occur within a cellular automaton, even though it is a "universe" with very few rules.

I started to look at the behaviors in Wolfram's rule 30 and rule 110 cellular automata because those have been selected as possible candidates for cryptographic systems. I wrote a program to evaluate patterns produced by these rules in a finite two dimensional field which wraps on both ends. I think I may have found other rules which may also prove promising for cryptographic use, but I need to do more evaluating. I'd ultimately like to use a three, four, or five dimensional system to see if I could build a useful PRNG but I will need to spend considerably more time ensuring that the patterns created by the rules remain complex. In the 2D system I evaluated, the vast majority of rules produces very predictable patterns.

See you when I get back.

A Steganography Scheme - Part 2

My text based steganography program is here! I completed the program yesterday, and Vanessa came up with a great name: Steganosaurus. Get in touch with me if you would like me to send you the program and the source code. I'm still considering publishing it on an open source hosting site.

You may recall my previous post about this steganography scheme, and I said I would tell you how it all works. So here goes:

My steganography program needs 4 pieces of information to embed or extract a secret message, it needs a file which will be converted, the base, the shift value, and a filename to which the converted message will be stored. The base and shift need a bit of explanation.

Base: All data on a computer is a number, and a number can be expressed multiple ways. I wrote about this in "A Steganography Scheme - Part 1". So the base tells Steganosaurus how to express the data. Should each number in the source file be converted into a series of values between 0-10, 0-50, 0-200? The choice is yours.

Shift: My program outputs a range of values from the source file, and each of them is between 0 and Base. How does this become readable text? That is the purpose of the shift, it is a value added to each piece of the converted file to make it into a character. So the result of the whole process is the contents of the original file expressed as numbers in the range Shift to Base + Shift. These numbers are converted into Unicode characters (UTF-8) so the end result is a readable file. You can see an example in my wiki entry about Steganosaurus. (It's probably easiest to see an example.) By using different shift values, you can hide the data from your original file in text from any language in the world.

So check it out, give it a shot, and ask me to send Steganosaurus to you.

A Steganography Scheme - Part 1

I've been thinking lately about steganography. Previous tools have hidden information in images, videos, sound files, and other large chunks of information. But I have not seen many examples of steganography with text (here is one). I decided to put together a simple and flexible text steganography tool just for fun.

The secrecy in a stegonography scheme rests in a shared algorithm between the sender and receiver which is not known by those trying to intercept the message and the fact that a hidden message is not apparent in the message being sent between the individuals. Anyone trying to find the hidden message faces two challenges:

1. How do I know when a hidden message is present?


2. How do I extract the hidden message?

Because I am explaining my stegenography idea to the world at large, item number two is no problem. I hope that the first question will not be so easily answered. In any case, you should not rely on this, or any other steganographic method to keep a secret. For that, you need encryption.

Now, on with the scheme!

I wanted users of my system to be able to send any kind of data by hiding it in a plaintext message. In order for the information to be hidden in text, it may have to be converted from one form to another. The same information can be expressed in multiple ways. For example, the number 14 can also be expressed as 1110 (binary), 16 (octal), and E (hexadecimal), depending on the encoding or base of the information. When information is converted into another form, someone who is decoding or reconstructing the information needs to know what form it is stored in. In many steganography schemes, the form of the data is constant and is a secret shared only by those who are supposed to be able to read the hidden message. I wanted to allow flexibility in the encoding, or alphabet, which is used to express the hidden message, so a message from my system will include a definition of the alphabet used to hide the secret message. Here are the steps for using my scheme:

1. Write the secret message


2. Choose an alphabet to encode with


3. Convert the secret message to the desired alphabet


4. Embed the converted secret message into an innocuous public message.

Now I've shared with you what it is that my steganography scheme will do. In my next post, I'll talk about how.