Probabilistic Behavioural Model Based on X-parameters

X-parameters are a powerful behavioural modelling formalism that extend the well-known scattering parameters to the nonlinear regime. However, uncertainties in the extraction of the value of the X-parameters may result in poor model behaviour. The aim of this study is to derive a Bayesian probabilistic X-parameter model showing a range of possible output values, corresponding to the typical variations seen between model and measurements due to random errors present during model extraction. In this work, a probabilistic posterior X-parameter model is developed for an RF Power Amplifier (PA) circuit which provides a probability distribution for each considered X-parameter value. Extra device measurement data can be used to update the posterior distribution, reducing the distribution variance and hence reducing the model uncertainty.