Daybrown, I see what you are saying, and maybe it would be true, if the signals were random and did not follow strict protocols established by modem handshake, and big bro had no access to your local phone company exchange (mamma bell for you I guess :P )
"SMPA is a real-time, passive and listening-only modem call monitoring tool. SMPA works bysimultaneously monitoring and analyzing the signals from both modems, and displaying the relevanttime-stamped information all in real-time. Test access is through DS1-level (T1 or E1) monitoringmode. Depending on platforms, a single SMPA can analyze one to many modem calls at the sametime" (Link)
A interceptor can distinguish between the two modems as you can see from the article from the link, even if you have shortwave radio hooked up to the modems. If it was so easy, no one would bother with encryption? Hmm, but I guess i don't have the expertise to say for sure. :-/ Looking in to this.
Edit: ok, with the 440 hz forks.. let's simplify: 2 people = identical japanese twins (A and B)
started to sing at same time on a phone line, how would you know who is singing what?
You are saying that this is the case bellow?
( signal coming from A ->
------------------------------->2 mixed signals (aka cacophany)-> |me intercepting| )
( signal coming from B ->
In reality I think that one signal is coming in from one end, and the other is coming from the other. You don't realy need to have inversion transformer to identify and separate the messages if you understand the language, it's all about the location in time/space.
( signal coming from A ->|me intercepting both at a phone company|<- signal coming from B )
Same with radio interception. That is the reason you need encryption. (aka jibberish)
cacophany is not a security solution :-S
Edit2: Addendum-
The 2 identical Japanese twins (modems) are actually not even singing at the same time random songs they are engaged in polite conversations with each other albeit in a incomprehensible language and at speeds that are to human ear like one continuous white noise, but to another 3d Japanese (modem) it is a conversation where both speakers have exactly(?) the same voice.