Assistance plotting radiation pattern

I am trying to make a polar plot using the following code yet it doesn't come out as smooth as desired. I'd appreciate some further guidance:

P = [-65,-67,-66,-68.2,-67,-67,-66,-65,-63,-62,-52,-42,-41.5,-47.39,-60.5,-61.7,-63.35,-67,-65.2,-66.1,-71,-67,-67,-68,-65,-65];
A = [-180,-165,-150,-135,-120,-105,-90,-75,-60,-45,-30,-15,0,15,30,39,45,60,75,90,105,120,135,150,165,180];
G = P+10;
polarplot(A*pi/180,abs(G));

Answer 1

[Star Strider]

I’m not certain what result you want.

One option is to interpolate your data:

Ai = linspace(min(A), max(A), 360);
Gi = interp1(A, G, Ai, 'spline');
figure
polarplot(Ai*pi/180,abs(Gi));

You will need to experiment with that to get the appropriate result.

[Yuval]

Thank you. I would naturally want to see distinct lobes, as befit a normal/standard radiation pattern. Is it possible with these data?

[Star Strider]

My pleasure.

I am not certain, since I do not know what sort of antenna (I assume an antenna) you are testing. Some antennas (such as verticals) are essentially isotropic, so there would be no lobes or only minor deviations from a uniform isotropic pattern.

My expertise with respect to antenna radiation patterns is limited to my experience as an amateur radio operator, not a communications engineer. I am specifically answering with respect to coding an approach to a solution to your problem.

[Yuval]

Yet is this the correct way to plot a radiation pattern? Assuming the data have lobes is the above code expected to plot them too?

[Star Strider]

It is. If your data have lobes, they will be plotted. There is some directivity, with a minimum at 0 and a maximum of 180.

[Yuval]

It seems the plot doesn't correspond to the radiation pattern of a standard horn antenna as shown in the attachment and is hence unfortunately wrong. Is the problem with the data or with the code rather? The P vector ought to be normalized by adding 41.5 to each of its elements and the plot should therefore be P+41.5 as a function of the angle. Any idea how to render a plot with greater semblance to that in the attachment?

[Star Strider]

I have no idea how you measured your data. Antenna radiation patterns usually are 3-dimensional, so the elevation is as important as the azimuth. Increasing the number of angles (resolution) at which you measured your data could be significant.

[Yuval]

The attachment shows a regular Cartesian plot of the same P+41.5 vs. the angle. As you can see the gain is not always positive, as in the polar plot. The gain is expected to be negative too, especially when normalized wrt the maximum power, viz. -41.5. Was this more helpful? Any ideas?

[Star Strider]

No ideas.

I was helping you with or original question, and plotting your vectors. Antenna theory is far from my areas of expertise.

The idea of ‘negative gain’ is essentially attenuation. This only makes sense if the units are dB, since negative in that sense simply means fractional.