patrick.melet@dmradiocom.fr wrote:
> hi
>
> I'm working with Matlab/Simulink to the system design and simulation of
> SNR at the output of the demodulator and AWGN in the channel.
>
> Then I'm going to make my models in VHDL language under Quartus and on
> a Stratix FPGA.

You may know, if you use Matlab/Simulink you can also generate DSP code
automatically.
Then your problem will be my problem: selection of sutiable adc card.
Because Simulink has restricted adc card support for automatic code
generation.

pls write me private messages. murselonder@yahoo.com

hi

I'm working with Matlab/Simulink to the system design and simulation of
SNR at the output of the demodulator and AWGN in the channel.

Then I'm going to make my models in VHDL language under Quartus and on
a Stratix FPGA.

Hi,

What are your tools? software and hardware I mean. I want to design PM
demodulator but my IF is 70 mhz. And I have been using C6713 DSK and
Matlab/Simulink. May we help each other.

patrick.melet@dmradiocom.fr wrote:

> hello,
> 
> I'm designing a receiver for FM demodulation and I have a little
> problem with the filters of the downconverter.
> 
[...]

> 
> My problem becomes from this last filter, when I take a bandwidth of 40
> kHz the ouput signal of my demodulator (after donwconverting I compute
> the derivative of the signal) is not my original 19,2 kHz sinus but
> it's a triangle
> and if I put a 80 kHz bandwidth my 19,2 kHz sinus is correct !!
> 
> What's happen ?

Unless there is a trivial mistake somewhere, it could be the effect of 
the excessive band limiting and/or nonlinear phase in the filter. Those 
effects result in the nonlinear distortion of the demodulated FM signal.


Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

http://www.abvolt.com

hello,

I'm designing a receiver for FM demodulation and I have a little
problem with the filters of the downconverter.

Here's my specifications : FM carrier 10,7 MHz, 20 kHz deviation.
I send to the modulator a 19.2 kHz sinus signal.

At the receiver, I first band-pass filter the carrier : center to 10,7
MHz with 80 kHz bandwidth.

Then I undersample at 4 MS/s : so I have got a first IF signal center
at 1,3 MHz with 80 kHz bandwidth.

Then I need to downconvert to 0 IF by multiplying with cos and sin to
obtain the I and Q signals.
So, I multiply with a sinus and a cosinus at 1,3 MHz and then I need to
put out the second order frequency by a low pass filter with 40 kHz
bandwidth.

My problem becomes from this last filter, when I take a bandwidth of 40
kHz the ouput signal of my demodulator (after donwconverting I compute
the derivative of the signal) is not my original 19,2 kHz sinus but
it's a triangle
and if I put a 80 kHz bandwidth my 19,2 kHz sinus is correct !!

What's happen ?

Any suggestions will be welcome...

Best regards.