Drs. Tarek El-Ghazawi and Volker Sorger have received a 3-year $675K award from the Airforce Office of Scientific Research

Professor Tarek El-Ghazawi (ECE), as the P.I, and Professor Volker Sorger (ECE), as the Co-PI , have received a 3-year $675K award from the Airforce Office of Scientific Research (AFOSR). The new project is titled RANC: A Residue Arithmetic Nanophotonic Computer. The goal of the project is to develop an integrated photonics computing system (from device to architecture) based on the residue number system (RNS) to achieve orders of magnitude improvements in computational speed per watt, over the current state-of-the-art, while again computing at the speed of the light. Residue arithmetic is of particular interest because it can represent a large number as a set of smaller numbers, which can be processed in parallel. Furthermore, RNS and nanophotonics have a natural affinity where most operations can be achieved as routing using electrically controlled directional switches (couplers), thereby giving rise to an innovative processing-in-network (PIN) paradigm. The project will explore a path for attojoule-per-bit efficient and fast electro-optic switching devices, and use them to develop optical compute engines based on residue arithmetic leading to multi-purpose nanophotonic computing. This grant falls along the same trajectory of the PIs vision to explore new PostMoore's law processor paradigms that can can bypass the technological obstacles presented by the end of the Moore's Law and Dennard's scaling by engendering new classes of efficient Nanophotonic processors. Since they started this pioneering direction four years ago, they have jointly secured multiple grants from NSF, ONR, AFOSR, and SRC totaling about $6.7M for this groundbreaking line of research to explore: photonic neuromorphic processors, reconfigurable processors and PDE solvers, residue number system processors, convolutional neural networks edge processors, and adaptive hybrid nanophotonic/plasmonic/electric networks-on-chip to support many core processors.