Encrypted messages arouse suspicion because they look like noise, and sending noise is suspicious.
Steganography allows for sending messages hidden inside other messages. So now I am not sending noise, but an innocuous message. Which could still arouse suspicion, but less so.
So sending encrypted messages is suspicious not just because it is noise, but noise in the contrast of the previous lack of noise (the void).
So one way would be to establish a continuous channel of background noise to the other party, and send encrypted messages to them when we need to. Of course, the establishment of a continuous channel of noise is still suspicious.
However, if this background noise was always present and assumed by our adversary, it would not be as suspicious. This is the “seeming void” of our adversary—Noise with zero perceived meaning, but through which we can exploit to send secret messages through.
So what is my seeming void? By definition, I don’t know. But could I?
Stego doesn’t work that way, sorry. The process is you take your plaintext and encrypt it, the reason is to whiten it (i.e. make it random in plain language). You treat that cyphertext as a bit/nybble sequence. You than take your transport target and (here is the big!!! requirement) take the LSB that should also be random and replace it with your cyphertext subsequences. A common payload is an image file. It really, really, really helps if the original file is not available to Mallet (other wise a simple diff will expose your subterfuge). Check out Bruce Schneier’s book Applied Cryptography 2ed (red cover) for a good intro, even if it is from the 90s. Neal Koblitz’s number theory books is also good but post-grad level. Just a suggestion from a Cypherpunk :)
Steganography of the seeming void.
Suppose I want to send a secret message.
Encrypted messages arouse suspicion because they look like noise, and sending noise is suspicious.
Steganography allows for sending messages hidden inside other messages. So now I am not sending noise, but an innocuous message. Which could still arouse suspicion, but less so.
So sending encrypted messages is suspicious not just because it is noise, but noise in the contrast of the previous lack of noise (the void).
So one way would be to establish a continuous channel of background noise to the other party, and send encrypted messages to them when we need to. Of course, the establishment of a continuous channel of noise is still suspicious.
However, if this background noise was always present and assumed by our adversary, it would not be as suspicious. This is the “seeming void” of our adversary—Noise with zero perceived meaning, but through which we can exploit to send secret messages through.
So what is my seeming void? By definition, I don’t know. But could I?
Stego doesn’t work that way, sorry. The process is you take your plaintext and encrypt it, the reason is to whiten it (i.e. make it random in plain language). You treat that cyphertext as a bit/nybble sequence. You than take your transport target and (here is the big!!! requirement) take the LSB that should also be random and replace it with your cyphertext subsequences. A common payload is an image file. It really, really, really helps if the original file is not available to Mallet (other wise a simple diff will expose your subterfuge). Check out Bruce Schneier’s book Applied Cryptography 2ed (red cover) for a good intro, even if it is from the 90s. Neal Koblitz’s number theory books is also good but post-grad level. Just a suggestion from a Cypherpunk :)
Thanks for your reply! What is “Mallet” in this context?