Two Computational Approaches for Noise Modeling of Advanced Microwave Transistors

A complete noise modeling of advanced microwave transistors is of great importance for the development of high sensitivity electronic systems. In the present paper, two computational approaches for modeling the noise performance of low noise microwave transistors are reported. The adopted techniques here presented involve the development of an artificial neural network tool and the implementation of an evolutionary algorithm. Each implemented approach is validated by a comparison between the expected values of the noise parameters (available from measurements) and the obtained data values. A thorough analysis of the results allows us to affirm that the application of these techniques quickly and originally solves the noise modeling task here examined.