Re: [Discuss-gnuradio] GMSK at 768kb/sec and other good stuff

[n4hy]

Right.  We did the "standard squelch" where we compare a "pitch 
frequency" bandpass filter's output  to a high frequency band pass 
filter's output  (same Q), and then compare the differences in the mean 
square output.  If the power levels are nearly the same, it is 
squelched.  If the low freq filter exceeds the high frequency filter  
output sufficiently and for a stable time,  we drop squelch.  There is a 
dead band region done by counting times above and times below and in the 
end, it worked immediately after a couple of bug fixes.  It isn't 
exactly hidden markov modeling of the state of the sound track but it 
has the advantage that it is immediate, cheap, and it works for voice 
communications.

Bob


Matt Ettus wrote:

> Robert McGwier wrote:
>
>  
>
> > What else did we forget?
> >    
>
> We did a proper squelch on the nbfm demod.
>
> Matt
>
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>
>