Vector Network Analyzer Sweep Power Measurement，and create the network!

[AceYoung938]

VNA_POW.txt
I use a program-controlled VNA to perform power sweep measurements. I set the software to measure S-parameters from 0 to -20dbm at a frequency of 2GHZ, with a total of 201 points.
The data format read by the software is an array of real and imaginary parts. Now I want to turn these data into network objects through the Scikit-rf library. How to do it? And these data are uncalibrated, how to calibrate?
Another question I have is how to use the scikit-rf library function to create a network object from the above type of file and draw its image. This file is the result file of sweep power measurement with a vector network analyzer in my laboratory. It is at 2GHZ frequency, the power range is -25dbm to 5dbm, the sweep power measurement result of the amplifier.
please help me

[cafeclimber]

Just to make sure I understand, you have already generated your .sNp files?

[AceYoung938]

No,Only the real part and imaginary part of the S parameter, each power point corresponds to an array.

[cafeclimber]

Oh sorry, just saw the upload txt file. Looking at that file, it IS indeed a .sNp file. Trying to load it myself, it looks like scikit-rf doesn't currently support Network objects with a power sweep instead of a frequency sweep.

A possible feature request @jhillairet

[AceYoung938]

thanks for your answer

[jhillairet]

I use a program-controlled VNA to perform power sweep measurements. I set the software to measure S-parameters from 0 to -20dbm at a frequency of 2GHZ, with a total of 201 points.
The data format read by the software is an array of real and imaginary parts. Now I want to turn these data into network objects through the Scikit-rf library. How to do it?

A Network is by definition a frequency-dependant data container as mentionned by @cafeclimber.

It is however possible to represent a power sweep dataset using a NetworkSet parametrized with the power values. Here is an example:

import numpy as np
import matplotlib.pyplot as plt
import skrf as rf

#%% Dummy data representing the real and imaginary parts of S11 for a power sweep
# from 0 to -20dbm at a frequency of 2GHZ, with a total of 201 points.
powers = np.linspace(0, -20, 201)
s11_r = np.random.rand(powers.size)
s11_i = np.random.rand(powers.size)

#%% Creating a NetworkSet containing 201 networks, each containing the S11 at 2GHz.
# Power values are defined using a named parameter 'power'
s11 = s11_r + 1j*s11_i
freq = rf.Frequency(2, 2, npoints=1, unit='GHz')

ntwks = [rf.Network(frequency=freq, s=s, params={'power': power}) for (s, power) in zip(s11, powers)]
ns = rf.NetworkSet(ntwks)

#%% Examples on how to access NetworkSet data
ns.sel({'power': [-5, -10]}) # -> NetworkSet for -5 and -10 dBm

#%% plot s vs power
_s = [n.s_db.squeeze() for n in ns]
_powers = [n.params['power'] for n in ns]

fig, ax = plt.subplots()
ax.plot(_powers, _s)
ax.set_xlabel('power [dBm]')
ax.set_ylabel('s11 [dB]')

Please refer to the NetworkSet Tutorial and this NetworkSet example.

And these data are uncalibrated, how to calibrate?

Please refer first to the calibration tutorial of the documentation and the calibration examples. Without additional information on your particular setup, it's hard for us to help us more.

Another question I have is how to use the scikit-rf library function to create a network object from the above type of file and draw its image. This file is the result file of sweep power measurement with a vector network analyzer in my laboratory. It is at 2GHZ frequency, the power range is -25dbm to 5dbm, the sweep power measurement result of the amplifier.

As mentionned by @cafeclimber , scikit-rf expects frequency dependant Touchstone file. Other kind of sweep are currently not supported.

[AceYoung938]

thank you，I will try your way now

[AceYoung938]

In #905 ， I created the SOLT calibration standard and got the calibration object 'cal'. As an example, the 'cal' object is created in the 10mhz to 6Ghz frequency range, can you calibrate the S-parameters of '_s' in jhillairet's code ？ @jhillairet