Compact parameterized macromodels for signal and power integrity analysis of RF and mixed signal systems

Linear macromodeling techniques are well established for compact dynamical modeling of complex signal and power distribution networks. In this work, we extend applicability of such reduced-order behavioral models to small-signal descriptions of complex circuit blocks typically found in RF or Analog and Mixed/Signal designs. In addition, we include in the models the explicit dependence on one or more design parameters, such as temperature, bias or gain, thus obtaining a multivariate small-signal behavioral macromodeling approach. The main outcome is a major reduction in the runtime required for transient system-level verification, which can be performed directly by simulating the surrogate macromodels rather than full transistor-level circuits. We demonstrate this approach through an Operarional Amplifier circuit block of a commercial 3G transceiver design.