S-parameter de-embedding

Hello all,

I made S-parameter measurements on some antennas we had made for near field materials characterization. Simple square patch microstrip antennas contacted by a SMA. I want to monitor the input impedance of the antenna at the ground plane. Perhaps a better way of saying it, the input impedance seen at the end of the SMA 50ohm t-line. Using SMA calibration standards sets the reference plane somewhere inside the SMA leaving a few millimeters of t-line before the ground plane. Using the s-parameters I already measured, is there a way to de-embed a length of lossless transmission line from the data?

Thanks in advance

De-embedding can be done in many ways.You can convert s-parameters of each block to ABCD parameters and then extract Transmission Medium s-parameters from global ones.Then the obtained new ABCD parameters should be converted again to s-parameters.

TDR (Time Domain Reflectometer) it's a very accurate method for de-embeding.

http://literature.cdn.keysight.com/l...990-9252EN.pdf

You just need to cascade a negative length of 50 Ohm line. Do you have an RF circuit simuator to do that?

I do not have access to a circuit simulator nor does our VNA have the time domain option.

There is a lot of info about how to convert to ABCD matrices for two port networks, but how would i go about doing it for a 1 port measurement?

For Antenna measurements, you can use 2 symmetrical face-to-face setup including identical transmission environment so you will obtain 3 2-ports s-parameters block.Like that..

Transmission Medium-1--->Antenna-1--->Air (short distance)--->Antenna-2--->Transmission Medium-2

Here there will be one unknown that's Air but it may be calculated/estimated.It's not an ideal case but it's worthy to try..