Measuring Mutual Coupling

Hello all,

I am testing two patches resonant at 5.25 and 10.5GHz. They are placed at about 1mm a part such that their feed are next to each other. Their dimensions are different (as evident)

Like in the figure. Now if I want to measure Mutual coupling. For first patch I should measure S12 and for second S21? using network analyser.

Am I thinking in the right direction?

Thanks in advance.

S21 and S12 must be the same

I was looking at this paper

IEEE Xplore - Sign In

It derives the foirmul afor mutual coupling using dimensions of the patch , Width and Radiation resistance of antenna. In my case Width of both are diffrent .. so what Width I should take, means of which antenna and why? similarly about Radiation resistance.

Secondly, the coupling is for E plane and H plane. Practical measurements would be by placing the antennas in horizontal and vertical plane?

Thanks in advance.

hi, if i use only one port to excite all my array elements,
so is there any issue of mutual coupling? S21 or S12 is the result of using 2 port feeding, but how bout if using only single port?

i also confius with the terms current density, current distribution and surface current, are they most likely the same?

i try to search for info at the internet, but i still cannot get concrete answer. please give ur comment, im greatly appreciate it. thank u

As far as my understanding is, when patches come close / next to each other there will be mutual coupling and the current densities will change , you can theoretically calculate it but am still seeking answer to how to measure them.

i think so too (when patches closer , mutual coupling is exist and significant) BUT many papers i read, all of them speaks on mutual coupling describe in S21..as we know, S21 parameter is from multiport applications....
so i totally out of idea, how to calculate/simulate those mutual couplinga s my array only have 1 input port..what more to say to measure them...( i wish to measure too)
theoretically calculate it? can you suggest me any related links/books/references ?

thanx a lot,

if i found anything useful to be share, ill post them here later =]

Correct. To measure a transmission from one antenna to the other, you need 2 ports.

Very simple. You remove the feed network, and instead connect port 1 directly to one antenna and port 2 directly to the other antenna. That's it. Very simple.

dear volker_muehlhaus:
thank for your kind reply....
so its really need 2 ports then..thx for the info...

about the use of simulation of antenna w/out feeding network...i get what you mean...
but, in my case ...my array have 8 elements with 1 single input port...

so what do u think?
erm, maybe i need to test each of the element alternately?
look forward fo ur reply. TQ

Change the model so that each patch has its own port. Now you can probe all the crosstalk values.

S21
S31
S41
...

wow...ok thanx a lot =]