kzt wrote:The frequency of a signal is directly related to how much data it can carry.
This is not true at all. With a massive signal to noise ratio you could send HD data over 60 Hz baseband, you'd just need:
(1080*1920*24*24) = 60*lg(1 + S/N) [Nyquist channel capacity]
Solving for S/N, the ratio is 2^19,906,560 - 1. Is that hard to obtain? Um, yeah, that's a practically noiseless system. But can such a system be sent on 60 Hz? Totally.
Besides, your statement becomes even more nonsensical when you take modulation into account. Almost every signal we send is modulated up to some high frequency, making the correlation between frequency and data capacity almost zero.
edit: A couple more things
Given that bandwidth of SR4 is high enough that it takes essentially no time to transmit enormous data sets, this suggests a fairly high bandwidth, as in many gigabits per second. So you are not going to be doing this on AM radio either, you are probably going to have to be in the high megahertz to gigahertz range. AKA microwaves.
When you're talking about actual signal processing, you really can't use the term "bandwidth" twice like that. The colloquial usage (as in, bits/second), is
not the signal usage, which is how wide the transmission frequency band is. What this sentence should read is:
"Given that
gross bitrate of SR4 is high enough...this suggests a fairly high
symbol rate, as in many gigabits per second."
That doesn't necessarily correlate to high bandwidth or even a high symbol rate -- you can just pump the crap out of signal power, or find some clever way to reduce noise power. It doesn't correlate to gigahertz bandwidth either, because you could just modulate it up the frequency spectrum arbitrarily high.