[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Linrad] Re: Feature Request

Hi Thomas,

> I want to look at a specific range in the waterfall, say from 4000Hz to
> 21000Hz.
> After a long frustrating period of clicking the -> <- arrows on the left
> and right upper corners of the display, I am unable to get the display
> in this range.  
I have no idea what you really want to do so my answer might
become a bit too complicated. More below.

> The up/down/left/right arrows do nothing.  I do not have
> a wheel mouse.
The arrows affect the baseband processing and can be used
when you do not have a wheel mouse. Nothing in the main waterfall
should be affected by the arrow keys.

> Is it possible to change one of the par* files to get this?

The main waterfall is governed by these parameters in par_xxx_wg
xpoints per pixel [12]
pixels per xpoint [0]
first xpoint [0]
xpoints [8184]

In case you want to use the wide waterfall as the primary information
about what happens in a certain frequency range you should set:
xpoints per pixel [N]
pixels per xpoint [1]

If you do not enable the second FFT use N=1. If you do use the
second FFT, make N the ratio between the transform sizes.

For the second FFT you have a parameter:
Second FFT bandwidth factor in powers of 2 [M]
The relation between M and N should be:
  M    N
  0    1
  1    2
  2    4
  3    8

This way you will get one pixel for each FFT bin in the waterfall display.
Now, if you want 4000 to 21000 Hz on a screen that is e.g. 1100 pixels wide 
you would want a bin separation of (21000-4000)/1100 = 15.4545 Hz/pixel.
Linrad can only compute FFT in sizes that are powers of two so you
would have to use an FFT size of 2048 or more. You may choose larger if you
like. The sampling rate that you would have to set would be:

  FFT size     Sampling rate

   2048            31.65 kHz
   4096            63.30 kHz
   8192           126.60 kHz

I would suggest a sampling speed of about 126 kHz which would mean that
the decimation should be 526. Not possible, but 525 will be ok:
You might use cic2=7, cic5=25 rcf=3

The sampling rate would become 126.984 kHz with 15.500992 Hz per pixel.
Set the first fft bandwidth to M*15 Hz. (M as above)

To make the first pixel = 4kHz in the waterfall, set first xpoint [258]
(you would get 3.999255 kHz for the first bin)

Set xpoints [1100] to make the waterfall use 1100 pixels. (The screen 
has to be a bit larger than that, the window is a little wider than
the waterfall.)

The last fft bin would be at 17.051 kHz.
> Or, add a box to allow input?
I see your requirement as rather odd. My guess is that there are 
better ways of doing what you look for. If you describe what
you are trying to do I might be able to give some hints.

The primary reason for having the main spectrum and waterfall is to
give the user the information required for proper operation of the
noise blanker. Typically the waterfall would have a large number
of bins for each pixel in which case the colour would correspond to
the average RMS power of strongest bin attributed to each pixel.
That procedure enhances narrowband signals (in case that is what
you are interested in the above procedure would not apply.)

It could be a good thing to allow the user to type in a start
frequency and a stop frequency. Since various things have to be 
ratios between integers, the result would not always be what
the user would expect though....

Leif / SM5BSZ

You received this message because you are subscribed to the Google Groups "Linrad" group.
To post to this group, send email to linrad@xxxxxxxxxxxxxxxx
To unsubscribe from this group, send email to linrad+unsubscribe@xxxxxxxxxxxxxxxx
For more options, visit this group at http://groups.google.com/group/linrad?hl=en