schrieb Patrick Strasser on 2011-01-25 16:55:
> schrieb Bernardo Gonçalves am 2011-01-25 10:15:
>> Basically, the problem is that it works just as mono, not stereo...
> Second, I looked at your flowgraph. You shift the difference part from
> center frequency 38kHz to 0 with the Xlating FIR filter, and then use a
> complex-to-float block. This will not result in the positive
> frequencies. You'd rather have to filter out the negative frequencies
> with a Hilbert transformer, that is a special FIR filter keeping only
> positive frequencies, before converting to float again.
Sorry, I was wrong. You did the low-pass filtering, that was already right.
But two other things I did notice:
1) The difference signal, which is shifted down from 38kHz is several
10dB lower than the sum signal. With different gains I got something
more stereo-like.
2) The sum signal is faster than the difference signal. In the
diff-signal path you have more blocks that are processed. Some delay
(delay block) could maybe fix this problem, but I did not get the right
value at a quick try.
Patrick
--
Engineers motto: cheap, good, fast: choose any two
Patrick Strasser <patrick dot strasser at tugraz dot at>
Student of Telematik, Techn. University Graz, Austria