Multidimensional time-domain macromodels for microwave applications

During the design of microwave circuits there is a significant need for efficient but accurate macromodels of components. These macromodels are derived from physical models of the components and may model either electrical or thermal behavior. However, increased levels of integration and higher power levels in microwave devices and modules have produced a need to include effects previously neglected during simulations. Accurate prediction of these effects involve solution of large systems of equations, the direct simulation of which is prohibitively CPU expensive. In this paper, an algorithm is proposed to form passive reduced-order macromodels of large linear networks that match the characteristics of the original network in time as well as other design parameters of the circuit. A novel feature of the algorithm is the ability to incorporate a set of design parameters within the reduced model.