EM-ANN MODELS FOR MICROSTRIP VIAS AND INTERCONNECTS IN DATASET CIRCUITS

A novel approach for accurate and efficient modeling of monolithic mic rowave/millimeter wave integrated circuit (MMIC) components by using e lectromagnetically trained artificial neural network (EM-ANN) software modules is presented, Full-wave EM analysis is employed to characteri ze MMIC components, Structures for simulation are chosen using design of experiments (DOE) methodology. EM-ANN models are then trained using physical parameters as inputs and S-parameters as outputs. Once train ed, the EM-ANN models are inserted into a commercial microwave circuit simulator where they provide results approaching the accuracy of the EM simulation tool used for characterization of the MMIC components wi thout increasing the analysis time significantly. The proposed techniq ue is capable of providing simulation models for MMIC components where models do not exist or are not accurate over the desired region of op eration, The approach has been verified by developing models for micro strip vias and interconnects in dataset circuits, A new hybrid (as) mo deling approach which makes use of existing approximate models for com ponents is introduced and shown to be a more efficient method for deve loping EM-ANN models, An example of using EM-ANN models to optimize th e component geometry is included.