Continue to Site

Welcome to

Welcome to our site! is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Calculating correlation between the E-fields of two antennas

Not open for further replies.


Newbie level 2
Feb 6, 2007
Reaction score
Trophy points
Activity points
Hi all,

I am having a bit of a problem computing the cross correlation between two Electric-field patterns of antennas. I will try to state my problem clearly:

I have E-field values of an antenna for each value of phi and theta.
The E-field values are complex numbers and are given as dB V/m.
So in Matlab I have a Matrix of dimensions 181(0-180 degrees values of φ) x 360(for 360 values of Θ).

Now I have another complex E-field matrix of the same dimensions giving the E-field radiation pattern of another antenna.

Now, I want to compare the two E-field patterns to see how much similar are they to one another. For this purpose I need to calculate their cross-correlation coefficient so that I get a single value between -1 and +1.

I am not sure on how to obtain the cross-correlation, because you cannot use the complex values directly into the correlation formulas, because then you get really wrong results.

I am also attaching the E-field matrices of the two antennas for your reference. These can be uploaded in Matlab using 'load' command.

Kindly help me in this regard, I will be very grateful.

Re: Calculating correlation between the E-fields of two ante

Comparing patterns and trying to collapse them into a single number has always been somewhat of a challenge. People routinely want to do that and I often find that it does not adequately describe an antenna or allow useful comparisons.

If I was trying to compare the two patterns you describe I would consider comparing normalized power patterns where I normalized to the highest E (converted to power) component. I would then plot nested 3D plots to see how they line up. Note that the power plots avoid the issue of comparing complex values.

I would also be inclined to use the conventional parameters of gain, beamwidth, sidelobe level, directivity and so on. The idea is to use parameters that people usually encounter.

I would be interested to see how your effort works out.

Good Luck.


    Points: 2
    Helpful Answer Positive Rating
Not open for further replies.

Part and Inventory Search

Welcome to