CAD Algorithms for Automated Nonlinear Macromodelling

University Of Minnesota-Twin Cities, Minneapolis MN

Investigators

Abstract

The research component of the proposal is to devise automatic macromodeling procedures that extract higher-level models from detailed lower-level ones for automated design of Systems-on-Chip. An important aspect of this process, particularly for analog, mixed-signal and RF modules, is the incorporation of nonlinearities. This is essential for distortion and intermodulation characteristics. This work is concerned chiefly with modeling weak circuit nonlinearities. A significane increase in the predicted fidelity over current (essentially linear theory-based) procedures is expected. The proposed work is based on Volterra non-linear series expansion that captures essential non-linearities of a broad class of devices which include power amplifiers, mixers, frequency doublers, and switched-capacitor filters. The work is likely to be expandable to companding filters and Analog-to-Digital converters.