Uncertainty Quantification for Nano-Scale Integrated Circuits and MEMS Design

Zheng Zhang
Seminar

CT

Argonne National Laboratory
Building 240/Room 4301
Host Emil Constantinescu
Event Sponsor Mathematics and Computing Science Seminar

Manufacturing uncertainties has become a major issue in today's semiconductor chip design. In order to improve the robustness of nano-scale chip design, it is highly desirable to develop novel SPICE-level simulators to efficiently quantify the uncertainties of integrated circuits (IC) and  microelectromechanical systems (MEMS). Mainstream simulators (such as HSPICE, Cadence Spectre and MEMS+) employ Monte Carlo for statistical analysis, requiring a huge number of repeated simulations. In this talk, I will present some novel stochastic spectral methods for stochastic circuit and MEMS simulation.
 
After an overview of existing uncertainty quantification techniques, I will present three main algorithms for fast stochastic IC and MEMS simulation. First, a stochastic spectral method called stochastic testing will be presented for fast stochastic DC, transient and AC analysis of nonlinear integrated circuits. This simulator can be regarded as a hybrid version of standard stochastic Galerkin and stochastic collocation methods. Second, based on stochastic testing, I will present an advanced periodic steady state simulator for efficient uncertainty analysis of analog/RF circuits (such as electronic oscillators). Finally, I will report a hierarchical uncertainty quantification approach that can handle complex systems with several subsystems and high-dimensional random parameters. The simulation results on various IC and MEMS examples have shown that our simulator can achieve 100x to 1000x speedup over state-of-the-art circuit and MEMS simulators.

About the speaker:
Mr. Zheng Zhang has been a PhD student in Electrical Engineering and Computer Science of MIT since Fall 2010. He received his M.Phil. degree from the University of Hong Kong in 2010, and his B. Eng. degree from Huazhong University of Science and Technology in 2008. His research interests include uncertainty quantification, tensor analysis, reduced-order modeling, and their applications to the computer-aided design of integrated circuits, MEMS, silicon photonics, and other engineering problems (e.g., power systems, MRI and data compression).
 
He received the 2014 Donald O. Pederson Best Paper Award from IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, the 2011 Li Ka-Shing Prize (university best MPhil/PhD thesis award) from the University of Hong Kong, the 2011 Mathworks Fellowship from MIT, and three additional best paper nominations at ASP-DAC2011, ICCAD 2011 and CICC 2014.