Research
Dynamically Adaptive circuits for Biosensing
Biosensing applications across different application spaces (health sensing, wearables, biometrics, bio-medical chips) all require high performance while consuming minimal power and area. This project explores dynamically adaptive circuits and systems for this space. The adaptability ensures that circuit/system behavior is modified only when necessary, thus limiting or eliminating additional overhead when the system does not require it.
Time-domain Neuromorphic AI Computing
Neuromorphic AI is still in the early stages of its evolutionary cycle. With energy efficiency and scalability as primary goals, we explore analog and mixed-signal neuromorphic blocks that use time delay as the primary signal for computing. This project involves integration of novel RRAM-type memory into the mixed-signal system.
HW Security
The security and trustworthiness of semiconductor ICs in the global supply-chain era is being increasingly emphasized. We look at organic analog/mixed-signal methodologies of introducing, detecting and mitigating HW trojans. We consider current analog test and verification methodologies, and exploit the inherent variations present in analog circuits to expose vulnerabilities.
Analog design in the CMOS+X domain
In this project, we explore new and upcoming devices and their applicability to analog design. How best can we leverage the advantages of these new devices to advance traditional analog design techniques? Our focus is on devices that are still compatible and/or integrable with modern CMOS processes.