Re: [Discuss-gnuradio] blks2impl/dbpsk.py question

[Stefan Brüns]

On Thursday 05 March 2009 19:12:15 Eric Blossom wrote:
> On Thu, Mar 05, 2009 at 05:37:00PM +0100, Dimitris Symeonidis wrote:
> > Hi List,
> > I was reading through the dbpsk.py block in blks2impl, and found
> > something that doesn't seem to make much sense (to me).
> > Lines 272-273 read:
> >         rot = 1
> >         rotated_const = map(lambda pt: pt * rot,
> > psk.constellation[arity]) Is this not equivalent to:
> >         rotated_const = psk.constellation[arity]
> > ??? I mean, it's just multiplying each output item of
> > psk.constellation with the number 1, or am I mis-reading something?
> >
> > Thanx for looking at this...
>
> I'm guessing that it's in there because at sometime someone wanted to
> rotate the constellation.  (If rot were a suitable complex value, the
> multiplication by rot would rotate it.)  rot = 1 means "don't rotate"

This may be needed based on the type of mpsk decoder in the chain. As 
currently there is a differential decoder, sybols will always be on positions 
(theta + i * 2pi/M; i=[1..M]; theta=0). With no differential decoder, it 
depends on the carrier tracking loops view of 'zero phase (theta)'. For the 
mspk_receiver_cc, the constellation points will be the same as stated for the 
diff. encoding. In general theta can have any value, and for the qpsk 
_encoder_ is set to 45 degrees.

Stefan

-- 
Stefan Brüns  /  Bergstraße 21  /  52062 Aachen
mailto:lurch at gmx.li  http://www.kawo1.rwth-aachen.de/~lurchi/
   phone: +49 241 53809034     mobile: +49 151 50412019