Statistical analysis of semiconductor devices

Abstract

A technique for the analysis of random dopant-induced effects in semiconductor devices is presented. It is based on the “small signal analysis” (perturbation) technique. It is computationally much more efficient than the existing purely “statistical” techniques, and it yields the information that can be directly used for the design of dopant fluctuation-resistant structures of semiconductor devices. This technique requires only the knowledge of variances of fluctuating doping concentrations and in this sense, it is a “second-moment characterization” technique. This technique can be naturally extended to take into account random fluctuations of oxide thickness and oxide charges in metal–oxide–semiconductor filed-effect transistor. The numerical implementation of this technique is discussed and numerous computational results are presented and compared with those previously published in the literature.