Title:  Analog and Neuromorphic computing with a framework on a reconfigurable platform

Committee: 

Dr. Hasler, Advisor       

Dr. Naeemi, Chair

Dr. Inan

Abstract:

The objective of the proposed research is to demonstrate an energy-efficient computing on a configurable platform, the FPAA, by leveraging the analog strengths, along with the development of a framework, to enable real-time systems on hardware. By taking inspiration from biology, fundamental blocks of neurons and synapses are built, understanding the computational advantages of such neural structures. To enable this computation and scale up from these modules, it is significantly important to have an infrastructure as well that adapts by taking care of the non-ideal effects like mismatches and variations, which commonly plague analog implementations. The programmability, through the presence of floating gates, helps to reduce these variations, thereby ultimately paving the path to take physical approaches to build larger systems in a holistic manner.