Thread Subject: Polar plot with dB scale

I am trying to plot radiation patterns for antennas on a
polar plot. Any ideas how I can convert the rho axis to a
log scale? Im using version 6.5

I want to know the answer as well, anyone can help?

I want to know the answer too, anyone help please?

In article <f976k4$j8k$1@fred.mathworks.com>,
Jennifer Cohen <j_co50@hotmail.com> wrote:
>I am trying to plot radiation patterns for antennas on a
>polar plot. Any ideas how I can convert the rho axis to a
>log scale? Im using version 6.5

You can't. If you did, then your plot would have to become infinitely
large in order to accomedate the polar pole which is always
at radius 0 (log of which is negative infinity.)
--
  "There are some ideas so wrong that only a very intelligent person
  could believe in them." -- George Orwell

Hi Roberson, thank you for replying so quickly! I think I
didn't state the problem clearly: it is the axis I want to
change, not the plot itself. Actually I want my plot's rho
scale to be (from out-side most circle in) 0dB, 10dB,
20dB,and 30dB, instead of the linear scale (from inside out)
0, 2, 4, 6, 8, 10. Thanks a lot for anyone helps, really
really appreciate it!!!

In article <fg5sso$se2$1@fred.mathworks.com>,
Chi Zhang <ela04cz@gmail.com> wrote:
>Hi Roberson, thank you for replying so quickly!

'Roberson' is my family name; my given name is 'Walter'.

>I think I
>didn't state the problem clearly: it is the axis I want to
>change, not the plot itself. Actually I want my plot's rho
>scale to be (from out-side most circle in) 0dB, 10dB,
>20dB,and 30dB, instead of the linear scale (from inside out)
>0, 2, 4, 6, 8, 10. Thanks a lot for anyone helps, really
>really appreciate it!!!

polar(1:10,1:10);
set(findobj(gca, 'String', ' 10'),'String', '0 dB');
set(findobj(gca, 'String', ' 8'),'String', '10 dB');
set(findobj(gca, 'String', ' 6'),'String', '20 dB');
set(findobj(gca, 'String', ' 4'),'String', '30 dB');
set(findobj(gca, 'String', ' 2'),'String', '40 dB');

To automate this more, you can find out what the strings
are that you need to replace:

yticks = get(gca,'YTick');
yticklabs = get(gca,'YTickLabel');
activelabs = cellstr(yticklabs(yticks>0,:));
numlabs = length(activelabs);
for K=1:numlabs
  set(findobj(gca, 'String', [' ' activelabs{K}]), ...
      'String', num2str(10*(numlabs-K),'%d dB'));
end


I don't promise that the pattern will always be two spaces followed
by the tick label; that was the pattern in the tests I did.
--
   "No one has the right to destroy another person's belief by
   demanding empirical evidence." -- Ann Landers

Hi,thank you so much for helping. I tried the code you
suggested, instead of giving flower petals like radiation
pattern, it gave me some spirals...I think it might just be
a small trick, as all I want to do is to change the linear
polar plot into semi log polar plot, just as people can
change x(linear)-y(linear)plot into x(linear)-y(log)plot, by
typing 'semilog'.I have checked website and help doc I got,
no answer found. Any ideas?

In article <fgfncq$pni$1@fred.mathworks.com>,
Chi Zhang <ela04cz@gmail.com> wrote:
>Hi,thank you so much for helping. I tried the code you
>suggested, instead of giving flower petals like radiation
>pattern, it gave me some spirals...

You said, "I want the axis to change, not the plot itself".
So the code doesn't change the plot, just how the plot is labelled.

If you are getting spirals and you want flower petals, then it
is likely that you need a log() taken at some point. So use the
code I provided and pass in the log() of your current rho data.


>I think it might just be
>a small trick, as all I want to do is to change the linear
>polar plot into semi log polar plot, just as people can
>change x(linear)-y(linear)plot into x(linear)-y(log)plot, by
>typing 'semilog'.I have checked website and help doc I got,
>no answer found. Any ideas?

No, it isn't just "a small trick". Matlab does not support
semi-log plots; as I pointed out in another posting in this thread,
doing so would require an infinite plot because of the infinity
of log(x) where x is 0 to 1. The best you can do is trick polar()
to relabel the concentric circles.

Within the last few days I have been in contact with Mathworks support
on the issue of using log() and polar() [I found a some bugs in the polar
plotting routines], and the official word from them is that there is
no support for log polar plots.
--
  "All is vanity." -- Ecclesiastes

Use mmpolar from the file exchange!

Markus

"Jennifer Cohen" <j_co50@hotmail.com> wrote in message
<f976k4$j8k$1@fred.mathworks.com>...
> I am trying to plot radiation patterns for antennas on a
> polar plot. Any ideas how I can convert the rho axis to
a
> log scale? Im using version 6.5