Measuring Noise Figure of Active Circuits usind S-Parameters files (.s2p) in ADS

[Teeago]

Hi everybody.

I'm modelling a receiver ina Agilent ADS using the S-Parameters files exported by our Network Analyzer from real circuit-blocks (Filter, LNA, etc).

The point is that the ADS calculate the NF of the filters, but don't the LNA one. In the frequency range that the lna's s21 goes above 0dB, its NF is shown as zero.

I'm attaching some pics to make my doubt more clear. The nf of the filter seems to be its s21 just inverted... but I don't if it's right or wrong.


Filter Response:


Filter's NF:


LNA's Response


LNA's NF (?):

Thank you.

 

[BigBoss]

For passive circuits NF is equal to insertion loss...That's why your insertion loss is at the same time your Noise Figure..
For circuits with active devices, it's not so true.From pure s-parameters, you can not simulate NF.The linear model must have either s-parameters with Noise Parameters or linear model must have intrinsic Noise elements in it..

 

[Teeago]

It's exactly what I was wary... Just needed more opinions.

Thank you, mate.