Multivariate macromodeling with stability and passivity constraints

Abstract

We present a general framework for the construction of guaranteed stable and passive multivariate macromodels from sampled frequency responses. The obtained macromodels embed in closed form the dependence on external parameters, through a data-driven approximation of input data samples based on orthogonal polynomial bases. The key novel contribution of this work is an extension to the multivariate and possibly high-dimensional case of Hamiltonian-based passivity check and enforcement algorithms, which can be applied to enforce both uniform stability and uniform passivity of the models. The modeling flow is demonstrated on a representative interconnect example.