Accurate temperature dependent noise models of microwave transistors based on neural networks

摘要

Recently, authors have proposed neural networks for modelling the temperature dependences of elements and parameters of microwave transistor small-signal equivalent circuit including noise. This neural model enables the prediction of modelled device noise parameters for any operating temperature. In this paper, an improvement of the neural model accuracy is proposed. It is done by using an additional neural network aimed to correct the noise parameters' values.