Development of extraction and optimization based large-signal models for thinned metamorphic high-electron mobility transistors on germanium

Abstract

HEMTs on germanium have the advantage that the substrate can be easily removed, which facilitates integration into low-cost MCM-D circuit implementations. Although germanium has (dispersive) characteristics similar to silicon, we show that the largesignal modeling of these thinned Ge based metamorphic high-electron mobility transistors (HEMTs) is similar to that of GaAs and InP HEMTs. Two types of look-up table based nonlinear models that are respectively based on direct extraction and optimization are developed and evaluated.

A new hybrid multibias analytical/decomposition-based FET parameter extraction algorithm with intelligent bias point selection

“…The parameter extraction approach that has been presented in this paper has attained the required level of robustness that allows it to efficiently process such large data sets. It forms an integral part of algorithms used in the generation of nonlinear models for the purposes of design and technology evaluation [9].…”

“…The aim of the selection algorithm is to find a relatively small number of bias points from which a consistent model representation can be extracted. Once accurate values for the extrinsic elements are known, they can be de-embedded from the complete set of multibias -parameters and a nonlinear model can be constructed as was done in [9]. The paper is concluded with a presentation and discussion of experimental results illustrating the accuracy and robustness of the enhanced multibias extraction algorithm.…”