Variability-aware circuit design techniques for silicon photonic circuits—Oct. 14

Professor Wim Bogaerts
Ghent University
 
Wednesday, October 14, 2020
11:00 AM 
Via Zoom: https://mit.zoom.us/j/95629349135

Variability-aware circuit design techniques for silicon photonic circuits

Abstract
Silicon Photonics is probably the only technology that really enables large-scale integration of photonic building blocks. This is made possible by the high-end manufacturing technology developed for the CMOS industry, combined with the very high refractive index contrast of silicon and silicon dioxide. This allows submicron waveguides that can be packed close together on a chip. However, the high index contrast makes the silicon waveguides also extremely sensitive to fabrication variations. When combining many elements in a circuit, this variability will affect the overall circuit yield, limiting the effective scale of the circuits. Therefore, circuit design for silicon photonics needs to incorporate the effects of fabrication variations early in the design process, optimizing the circuit layout to maximize the yield. We will discuss these effects and the workflows needed to implement variability aware design for silicon photonics.

Biography
Wim Bogaerts is a professor in the Photonics Research Group  at Ghent University and IMEC. He completed his PhD in 2004, pioneering the use of CMOS tools to make photonic circuits. Between 2000 and 2010, he was instrumental in the buildup of IMEC’s silicon photonics technology. In parallel, he also started developing the design tool IPKISS  to implement complex silicon photonic circuits. In 2014, he co-founded Luceda Photonics, bringing his design tools to the market. In 2016 he returned full-time to Ghent University and IMEC on research grant of the European Research Council. His research now focuses on the challenges for large-scale photonic circuits and the new field of programmable photonics. He is a senior member of IEEE, OSA and SPIE.

Organized by the Research Laboratory of Electronics (RLE) and Department of Electrical Engineering & Computer Science (EECS)