Results 1 to 1 of 1
I've been studying up XOR quite a bit, for various reasons. Mostly a trip back into the world of programming, but I've sort of been challenging myself. See, I learned ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
- 05-07-2008 #1
Can you really make XOR secure?
I've been studying up XOR quite a bit, for various reasons. Mostly a trip back into the world of programming, but I've sort of been challenging myself. See, I learned relatively quickly that XOR wasn't really encryption, it was just a tool used for it, and most of the tutorials that I saw were simply Vigenère ciphers in disguise. I then quickly found out that XOR operations are used in many strong ciphers, such as RC4. Well, I challenged myself to preserve the simplicity of XOR in the Vigenère configuration, but to address all of the security flaws that it held. People have done this before me of course, and I borrowed some of their hardeners, and then added my own to address the weaknesses with his system. Basically, I won't get into the details, but I addressed every researchable weakness I could find, but I'm not cryptanalyst and I can't really speak for the true strength of it. I did fix one nasty little thing that made xor encryption notoriously insecure, and that is that if you xor'd your encrypted data with the plain data, the key would be revealed. In my configuration ( I want to avoid actually calling it a cipher ), this is checked, but mostly through a form of redundancy.
Now, going through all of that, I was wondering if anyone here really knows much about it. Anyone ever tried doing the same thing, think it's really possible? I figured now would be the best time to reach out to any people with better cryptoanalytical(sp?) skills too. After all, I've put a lot of time into creating the tool, now I really want to see if it's safe to use it.