We want to use light to make the world a better place. We combine optics with a few different technical fields, including nanotechnologies, high performance cloud computing, signal processing, and AI, to explore novel applications in imaging, computing, and energy. We also study fundamental optical sciences simply because they are interesting.
Multi-modal light-sensing pixels
Light-sensing pixels in a camera only detect the brightness. Information such as wavelength, incident angle, polarization of light are all lost. Exploiting unique interactions between light and nanostructures, we are building next-generation coherent pixels to retreive multi-modal light field information. Imaging chips built upon coherent pixels extract task-specific information most efficiently. Leaveraging upon the existing CMOS sensor architecture, these chips are low cost and can become the building element for pervasive visual sensing
Computational electrodynamics, cloud and neural network
Future photonic devices will heavily rely on optimization. Cloud computing, with its unlimited on-demand computing resources, is the ideal platform for high throughput optimization.For example, the frequency scan in Rigorous coupled-wave analysis (RCWA) method can be done in minutes in the cloud instead of hours or days. We are developing cloud-ready algorthims as well as neural network models for next-generation optimization engines.
Quantum electrodynamic simulator
We are developing computational tools to model classical-quantum hybrid devices. It combines classical and quantum electrodynamics simulators. We use the tool to investigate quantum antennas, quantum photonic crystals, quantum metasurfaces, all made from two-level systems such as quantum dots and atoms.
Imaging beyond the limit
Space is the enemy of imaging. When light passing through a long space, finer details of a scence are lost. In electrical engineering, we call the space a low-pass filter. Through spatial frequency mixing, we manage to use low frequencies to carry high-frequency details. Combined with compressive sensing, it allows us to see objects smaller than wavelength from a far-away distance.
Human-computer interface for technical computing
Exploring smart interface assisted by artificial intelligence and new interaction models for example augmented reality, we want to make full-wave electrodynamic simulation much more easily accessible. We think that today's simulation technolgies are too complex, requiring skills such as CAD and knowledge of boundary condition that are completely irrelevant to the physics. Our goal is to allow anyone with high school education to run useful electrodynamic simulations.
Radiative cooling technologies
Sky is cold. Why not use it to cool our workspaces, powerplants, and data centers? The challenge is that sky is too far away. But it is not a long distance away for light. We are developing self-cooling technologies that convert heat to light and send it to the sky. Exploiting radiating concentration, we make such cooling 10X better.