Research

Current Research Focus

Heterogeneous integration: Multi-domain modeling, simulation and prototyping of heterogeneous systems, driven by advanced packaging technology and chiplet design for energy-efficient AI computing.
AI acceleration for intelligent systems: Integration of CMOS, emerging logic and memory devices, non-conventional circuits and architectures to accelerate on-chip machine learning, with a particular interest in dynamic systems and real-time applications.
Neural-inspired computing: Joint algorithm-hardware exploration inspired by neurophysiology at the device, circuit and architecture level, in order to achieve fundamentally new learning capabilities with high energy efficiency.
Reconfigurable systems: Programmable design of circuits and systems to achieve a new platform for high-performance computing and design reconfiguration. The applications expand from emulation and prototyping to application-specific acceleration.

Predictive Technology Model (PTM): Accurate, customizable, and predictive model files for transistor and interconnect technologies. These predictive model files are compatible with standard circuit simulators and scalable with a wide range of process variations. From 2005 to 2012, PTM created model files for bulk CMOS until the 22nm node, FinFET (double gate) device down to the 7nm node, and carbon nanotube device.
Cross-layer design for resilience: Modeling, simulation, characterization, and resilient design solutions. The goal is to significantly improve the design quality, efficiency, and predictability of nanoscale digital and AMS circuits, with guaranteed tool-to-hardware matching.

We would like to express our sincere gratitude to the following sponsors for their generous support and assistance in our research projects:

Defense Advanced Research Projects Agency (DARPA), Department of Energy (DOE), National Science Foundation (NSF), and Semiconductor Research Corporation (SRC)
Fermi National Accelerator Laboratory, Oak Ridge National Laboratory, Pacific Northwest National Laboratory, and Sandia National Laboratories
Boehringer Ingelheim, Charles Stark Draper Laboratory, Cisco, GlobalFoundries, IBM, Intel, MaxLinear, Qualcomm, Samsung, and Texas Instruments