Re: [Discuss-gnuradio] 802.11 and Bluetooth

[David P. Reed]

The USRP contains 2 A/D's that can run at 64 MSPS each.   That means 
that it is quite possible to get 32 MHz bandwidth Nyquist sampling at 10 
bits/sample from each A/D.  If you wanted to, you could split the band 
into two parts, giving 64 MHz bandwidth.  That level of sampling would 
require 320 Mb/s or 640 Mb/s for two bands.

The limit is due to the USB path to the computer, which is neither 
compressed (10 bits of sample sent in larger word sizes) nor run at full 
rate (which would be 400 Mb/s if it were possible to run flat out).

But there is another point to be made about sampling.   Even though you 
can't sample the full 80 MHz bandwidth of Bluetooth without aliasing, it 
turns out that you don't need to care about aliasing when decoding 
Bluetooth that much.   Transmitting is a different case, but if you 
think about it, all that aliasing causes is for some noise to be added 
in from the aliased subchannels.

So you don't need full Nyquist-rate sampling for  decoding Bluetooth's 
80 MHz bandwidth, if you have a high enough SINR.

Bluetooth is only about 1 Mb/s data rate, so you should be able to 
decode it with only a few MSPS of sampling.