학술논문

QuickSim: Efficient and Accurate Physical Simulation of Silicon Dangling Bond Logic
Document Type
Conference
Source
2023 IEEE 23rd International Conference on Nanotechnology (NANO) Nanotechnology (NANO), 2023 IEEE 23rd International Conference on. :817-822 Jul, 2023
Subject
Bioengineering
Components, Circuits, Devices and Systems
Engineered Materials, Dielectrics and Plasmas
Fields, Waves and Electromagnetics
General Topics for Engineers
Photonics and Electrooptics
Power, Energy and Industry Applications
Robotics and Control Systems
Fabrication
Industries
Machine learning algorithms
Design automation
Runtime
Statistical analysis
Layout
Language
ISSN
1944-9380
Abstract
Silicon Dangling Bonds have established themselves as a promising competitor in the field of beyond-CMOS technologies. Their integration density and potential for energy dissipation advantages of several orders of magnitude over conventional circuit technologies sparked the interest of academia and industry alike. While fabrication capabilities advance rapidly and first design automation methodologies have been proposed, physical simulation effectiveness has yet to keep pace. Established algorithms in this domain suffer either from exponential runtime behavior or subpar accuracy levels. In this work, we propose a novel algorithm for the physical simulation of Silicon Dangling Bond systems based on statistical methods that offers both a time-to-solution and an accuracy advantage over the state of the art by more than one order of magnitude and a factor of more than three, respectively, as demonstrated by an exhaustive experimental evaluation.