Time-domain finite-element reduction-recovery methods for large-scale electromagnetics-based analysis and design of next-generation integrated circuits

Houle Gan, Purdue University

Abstract

The past few decades have witnessed an increased need of full-wave electromagnetics-based analysis of integrated circuits for four main reasons: (1) reduced feature sizes that lead to sub-wavelength optical lithography, (2) increased operating frequencies that increase full-wave effects, (3) increased level of integration, and (4) the advent of alternative technologies such as on-chip wireless communication. However, the analysis and design of integrated circuits (ICs) imposes many unique challenges on electromagnetic analysis such as large problem size, small physical dimensions, large aspect ratio, high conductor loss, and strong non-uniformity. State-of-the-art computational EM algorithms require O( N) or O(NlogN) computation with O(N) memory (N being the number of unknowns). For the next generation IC design problem, however, even O(N) is prohibitively expensive. In the first part of this work, time-domain finite-element reduction-recovery methods are developed to overcome the large problem size of O( N) to achieve a complexity of O(M), with M<

Degree

Ph.D.

Advisors

Jiao, Purdue University.

Subject Area

Electrical engineering

COinS