spectrum imbalance

"XDY" <yuxdll@pacific.net.sg> wrote in message
news:cl38pi$b98$1@nobel.pacific.net.sg...
> Hi friends,
>
> I am developing an algorithm which is very sensitive to the shape of input
> signal spectrum. In ideal situation, singal spectrum should be
symmectrical
> about zero frequency. however, the real signal captured from IF output is
> imbalanced. I guess there is something wrong in RF downconversion
circuits.
> So, I feed a white noise to the RF input, and capture the IF output again,
> it showed a sloped spectrum instead of a flat spectrum in the span of
> +-10kHz. It may be due to RF filter ripple. But when I feed in a CW to
> measure the spectrum from spectrum analyzer, there are no changes in the
> frequency span.

Explain this last part in more detail. You take a CW signal, send it through
your system and look at it using a spectrum analyzer - right? Wouldn't you
have to 'sweep' or 'step' your CW signal across the BW of interest to see
what the shape of the spectrum looks like on the spectrum analyzer ( you'd
need to setup the analyzer to 'peak hold' the values on a trace). If you
look at only 1 CW signal, you are only looking at 1 point in the frequency
spectrum.

Cheers
Bhaskar

>
> Is there any one has idea which impairement in RF receive chain can cause
> the slope spectrum?
>
> Best Regards,
> XD
>
>

Hi friends,

I am developing an algorithm which is very sensitive to the shape of input
signal spectrum. In ideal situation, singal spectrum should be symmectrical
about zero frequency. however, the real signal captured from IF output is
imbalanced. I guess there is something wrong in RF downconversion circuits.
So, I feed a white noise to the RF input, and capture the IF output again,
it showed a sloped spectrum instead of a flat spectrum in the span of
+-10kHz. It may be due to RF filter ripple. But when I feed in a CW to
measure the spectrum from spectrum analyzer, there are no changes in the
frequency span.

Is there any one has idea which impairement in RF receive chain can cause
the slope spectrum?

Best Regards,
XD

Hi friends,

I am developing an algorithm which is very sensitive to the shape of input
signal spectrum. In ideal situation, singal spectrum should be symmectrical
about zero frequency. however, the real signal captured from IF output is
imbalanced. I guess there is something wrong in RF downconversion circuits.
So, I feed a white noise to the RF input, and capture the IF output again,
it showed a sloped spectrum instead of a flat spectrum in the span of
+-10kHz. It may be due to RF filter ripple. But when I feed in a CW to
measure the spectrum from spectrum analyzer, there are no changes in the
frequency span.

Is there any one has idea which impairement in RF receive chain can cause
the slope spectrum?

Best Regards,
XD