An exclusive raffle opportunity for active members like you! Complete your profile, answer questions and get your first accepted badge to enter the raffle.
For electrically large problems it is often more computationally efficient to represent antennas, especially antenna arrays, with point sources. Could someone some light on the mismatch problem experienced with point sources?
Consider two patch antennas some distance apart. The top patch, with regards to its boresight direction, is directed 180 degrees away from the bottom patch.
We are interested in computing the coupling between the two antennas. In the first case we will compute the coupling using the exact geometry for the antennas for reference purposes. In the second case we will replace the antennas with point sources.
Computing the coupling for the full model using an S-parameter request yields an S21 value of -47.16 dB.
To use the point sources, we first solve a single patch and for the far fields export the *.ffe file. Then a new model is created whereby a radiation pattern point source is set up for the bottom patch, and a radiation pattern point receiver for the top patch. For both the source and receiver, the *.ffe file previously exported is imported.
The transmit and receive power in the point sources are given in the OUT file or available under 'Power' in POSTFEKO:
We see the radiation pattern point sources report some 10 dB stronger coupling than the full wave model (S-parameter S21). The difference is due to mismatch: the S21 parameter takes into account the mismatch losses, but the radiation patterns do not.
We can include the mismatch losses in the coupling for the radiation patterns as follows:
All of the above calculations can be performed directly in POSTFEKO using the 'Add Math' functionality on the result palette.
The example files are available ().
Unable to find an attachment - read this blog
