ADS LNA NF simulation question.

I got problem when I simulate the noise figure of my LNA design, I always get a unreasonable nf(1). Why? Is there something wrong in my simulation setup?

The above simulation result is the NF of an old CMOS LNA, you can see the nf(1) (or m2) is too large to be reasonable. The NF measurment result is around 3.8dB.

This is the simulation result of another LNA design, again the nf(1) is surprisingly large. This LNA, which has a very common structure, is shown below:

Summary of this LNA simulation result:

And the smith chart with NF and GA (available gain) circles is:

The definition of nf(k) in ADS is:

Noise Figure
The parameter nf(k) in the dataset is the noise figure at output port k. You can view results using the Data Display. The noise figure for each port will appear in the variables list as nf(1), nf(2), and so on. nf displays noise figures for all ports.

When noise figure is calculated at a port, the other ports in the network are terminated in their respective impedances.

For a 2-port circuit, the noise figure is the signal-to-noise ratio at the input, divided by the signal-to-noise ratio at the output. It has default units of dB. For a circuit with more than two ports, the noise figure is the ratio of the total noise at the output port to the transmitted input noise. The total noise is the transmitted input noise plus the noise contributed by the network. The transmitted input noise represents the portion of the incident thermal noise (kTB, where k=Boltzmann's constant, 1.381...x10-23 J/K, T=290 K, B=1 Hz) which passes through the system.

申明：网友回复良莠不齐，仅供参考。如需专业帮助，请学习易迪拓培训专家讲授的ADS视频培训课程。