A neural network model of silicon-based millimeter-wave coplanar waveguide

In this paper, neural network modeling techniques are presented for millimeter-wave modeling of silicon-based millimeter-wave coplanar waveguide. The neural network is trained to learn the mapping between the geometrical variables and S parameter of the coplanar waveguide. Once trained with the EM data, this model provides accurate and fast prediction of the measurement data of differential CPW with geometry parameters as variables. Experiments in comparison with input-output relationships by the proposed neural network model and measurement data are included to demonstrate the merits of this new model.