DoE Phase II
Distinguished Professor S. J. Ben Yoo (Electrical and Computer Engineering) wins a new team project led by MIT to demonstrate a Phase-change Reconfigurable Optical WavEfront Synthesis System (PROWESS). PROWESS comprises metasurfaces made of phase-change materials (PCMs), where their phase transition and hence refractive index are electrothermally tuned via Si micro-heaters. Switching of a large metasurface matrix is facilitated by integration with Si CMOS transistor arrays. The team will further explore two emerging applications uniquely enabled by PROWESS by implementing a reconfigurable optical neural network (ONN) and a high-information-density infrared imaging system. The reconfigurable ONN can not only adapt to different computation tasks on-the-fly but additionally allows dynamic training for instance via back propagation, dramatically streamlining the deep learning process. In the latter example, PROWESS enables real-time collection of multi-dimensional information including light wavelength, polarization, and incident angle, empowering imaging systems with unprecedented intelligence capabilities. The PROWESS team consists of Juejun Hu (lead PI MIT), S.J.Ben Yoo (UC Davis), Myungkoo Kang (U Central Florida), Tian Gu (MIT), Hualiang Zhang (UMass Lowell) supported by NSF ASCENT program at $1.5M over three years.