# Forum: GNU Radio Problem in simulation on GRC at the measure of BER in digital modulations with Chunks to Symbol in r

Announcement (2017-05-07): www.ruby-forum.com is now read-only since I unfortunately do not have the time to support and maintain the forum any more. Please see rubyonrails.org/community and ruby-lang.org/en/community for other Rails- und Ruby-related community platforms.
on 2014-05-14 05:26
Attachment: Mpsk.grc.png (200 KB)
Attachment: 8psk-2db-ber.jpg (80 KB)
Attachment: 8psk-8dB-ber.jpg (80 KB)
Attachment: 8psk-12db-ber.jpg (70 KB)
Attachment: 8psk-ber.jpg (400 KB)
```Good Evening,

I am a student of eletronic engineer in Brazil and I am using GNURadio
Companion as a research.
I have attached some pictures of my model on GRC and the problems.
In this case, I made a circuit with PSK modulation, using the block
Chunks to Symbol.
Before and after the the Channel Model, I used the root raised cosine
filter to get a better signal of FFT at the WX GUI FFT Sink.
Using the software called octave I could measure the power at the output
of filters, First I read the file saved from the block file sink and
after I could measure the power using the equation.
mean(abs(in).^2).
After this process i could see that the power at the output of Chuns to
Symbol was equal the number 1 and in the output of the first filter was
very small number, different of 1 and the factor that made this happen
was the interpolation factor in the properties of the root raised cosine
filter.
I couldn't find a relation of the interpolation factor of the filter to
get the number of power equal the value at the output of Chunks to
symbol.
After some tests i get a result equal the number 1 in the output of the
two filters, but with a value very big of noise the system measured a
value of BER equal zero and the constellation was very clear.
When I use this values of the model attached I can find results very
near of the theorical values from the curve Eb/N0 plotted at matlab, but
some samples are different of the theorical value.
My doubt is, how can I do to make the BER equal the value theorical(
from the curve of EB/N0 of the modulation)and get a good FFT signal ?```
This topic is locked and can not be replied to.