Study on the Yield of Self-Healing Carbon Nanotube/Nanowire-Based System

Abstract

This thesis is a study on the yield of a digital computing system built with an emerging technology, carbon-nanotube/nanowire. The emphasis is given on the yield improvement. This thesis presents an extensive and comprehensive review of the recent practices of carbon-nanotube/nanowire technology. The objectives of this thesis are: characterization of a carbon-nanotube/nanowire array-based computing architecture; architecture-based characterization and probabilistic parameterization of practical defects and failures; and incremental modeling and analysis of the yield of the system with such architecture. Based on the proposed yield model, parametric simulation is conducted to reveal the synergistic impact of conventional redundancy-based defect/failure-tolerance and self-healing capability on the yield through comparative study.