CW NF

CW NF - Noise Fig in CW mode.

Once the steady-state response of the device is determined, the noise analysis can be treated as the perturbation of the steady-state solution through the linearization of the equilibrium point. The principle of the noise analysis is:

1. Turn off all noise sources and calculate the large-signal steady-state solution.

2. Linearize the circuit around steady-state solution, apply the perturbation, and calculate the output noise.

The conversion matrix is implemented in VISION.

To summarize this method, consider a two-port device, as shown in the following figure :

Assuming the steady-state of the two-port device has been determined, it can be

described as follows:

Note that in general I2 (f k ) at a frequency f k depends on V1 (f p ) and V2 (f p ) for all harmonics (p =0,1,2...,K).

Then the noise will produce a small displacement of the steady-state solution as:

Developing the Taylor series to the first order yields the following:

Consequently, the current variation δI 2 (f k ) for all the harmonics is represented in a matrix form as:

C is the conversion matrix. This matrix represents the current sensitivity on the voltage variation. by interconnecting the conversion matrix of all the device, the noise charcateritics iof the circuit can be obteined by a linear analysis.

Then it is possible to calculate the ratio for each fk frequency:

In VISION software only the fundamental f0 frequency is analysed.