Artificial neural network models for the bend discontinuities in striplinecircuits

Stripline discontinuities are basic elements of many stripline circuits, su ch as the multilayer microwave monolithic ICs and the interconnect systems in highspeed digital circuits. Bend as one type of the stripline discontinu ities will be modeled in this paper. A multilayer perceptron neural network(MLPNN) is used to model the bend dis continuities in stripline circuits. The MLPNN is electromagnetically develo ped with a set of training data that are produced by the full-wave finite-d ifference time-domain (FDTD) method. The full-factor design of experiments is used to determine the size of the training data.