Since your constellation is complex, then its dimensionality (in the
chunks_to_symbos block, as well as in the PCCC decoder block)
should be set to 1 (you are now have it at 2).
The convention is that the chunks_to_symbols array is arranged as
a11, a12, … a1d, a21 a22,…,a2d,… aM1, aM2, …aMd
so that it is to be understood as the 2 D array
a11, a12, … a1d,
a21 a22,…,a2d,
…
aM1, aM2, …aMd
with dimensionality d.
Then, at each instance, it outputs the “row” that is indexed by the
input
symbol.
Achilleas
PS: I also suggest you make a variable “constellation” and put in there
your constellation and then input it in both
the chunks_to_symbols AND the PCCC decoder block so that you are sure
you
are consistent.
Hi
Thank you for your help.
Now im trying to use only the pccc encoder and pccc decoder blocks, without modulation. Im using the awgn1o2_8.fsm for the fsm parameter because i need a state
machine with those characteristics, and im using the error rate block to make sure that the source is the same than the output of the decoder. But this value is in the range between 0.02 and 0.06 (when the source is a random source) when it should be 0 (im not using any noise or channel
blocks) and is 0.02 with the vector source V = (0, 1, 0,1,1,0,0,1). Does
the
blocks itself introduce an error?
Also i would want to know if i can create my own fsm file in the case i
need a state machine with characteristics I=2, S=64,O=8 (There is any in
the
fsm_files folder). I created one but i don`t know either which is the
format
i should use nor where to save it.
How the Bits per Symbol parameter influences in the error rate?
Again i apologize for my english.
Thank you very much for your help.
FSM284_Vector_Cod_decod_turbo_error_rate.grc
http://gnuradio.4.n7.nabble.com/file/n53264/FSM284_Vector_Cod_decod_turbo_error_rate.grc
–
View this message in context:
http://gnuradio.4.n7.nabble.com/pccc-encoder-tp53173p53264.html
Sent from the GnuRadio mailing list archive at Nabble.com.