williamch wrote:

> Hi all,
>     I have implemented FM demodulation(including Mono and Stereo) on FPGA.
> When it work on Mono mode, the audio quality sounds good. But if I changed
> the mode to the stereo, the audio quality sounds not very well. I have
> checked the spectrum of m(t), and there are high noise floor over high
> frequencies(looks like the response of a high pass filter).

But of course.
If you use the discrimimator with the differentiation of the phase, the 
resultant noise will look like that.


  Even I changed
> the differentiator from [1 -1] to atan2(z(n)*conj(z(n-1)), the result is
> same.

No wonder.

  The phase of pilot tone looks like ok, because the separateness of
> L-R channel is fine. What could be the major cause to make this kind
> noise? What/How can I do?

Most of the noise is probably related to the quantization artifacts of 
your processing. Take a close look at the precision loss in the 
computations.

>     My demodulator work on Fs = 256k. The NCO used for IF Mixer and
> downconvertor(8.192M(Real)->2.048M(Complex)). 
>     If you can give me some suggestions, I will be very appreciate your
> kindly help.

Your appreciation means too much for me. $100 will be just all right.


Vladimir Vassilevsky

DSP and Mixed Signal Design Consultant

http://www.abvolt.com

williamch wrote:
> Hi all,
>     I have implemented FM demodulation(including Mono and Stereo) on FPGA.
> When it work on Mono mode, the audio quality sounds good. But if I changed
> the mode to the stereo, the audio quality sounds not very well. I have
> checked the spectrum of m(t), and there are high noise floor over high
> frequencies(looks like the response of a high pass filter). Even I changed
> the differentiator from [1 -1] to atan2(z(n)*conj(z(n-1)), the result is
> same. The phase of pilot tone looks like ok, because the separateness of
> L-R channel is fine. What could be the major cause to make this kind
> noise? What/How can I do?
>     My demodulator work on Fs = 256k. The NCO used for IF Mixer and
> downconvertor(8.192M(Real)->2.048M(Complex)). 
>     If you can give me some suggestions, I will be very appreciate your
> kindly help.

Are you on the edge of adequate sifnal strength? Stereo costs 3 dB.

Jerry
-- 
Engineering is the art of making what you want from things you can get.
¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯¯

Hi all,
    I have implemented FM demodulation(including Mono and Stereo) on FPGA.
When it work on Mono mode, the audio quality sounds good. But if I changed
the mode to the stereo, the audio quality sounds not very well. I have
checked the spectrum of m(t), and there are high noise floor over high
frequencies(looks like the response of a high pass filter). Even I changed
the differentiator from [1 -1] to atan2(z(n)*conj(z(n-1)), the result is
same. The phase of pilot tone looks like ok, because the separateness of
L-R channel is fine. What could be the major cause to make this kind
noise? What/How can I do?
    My demodulator work on Fs = 256k. The NCO used for IF Mixer and
downconvertor(8.192M(Real)->2.048M(Complex)). 
    If you can give me some suggestions, I will be very appreciate your
kindly help.